This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
 Author's personal copy

Earth-Science Reviews 108 (2011) 16–33

Contents lists available at ScienceDirect

Earth-Science Reviews
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r s c i r ev

Secular trends in the geologic record and the supercontinent cycle☆

Dwight C. Bradley ⁎
U.S. Geological Survey, 4200 University Drive, Anchorage, AK 99508, USA

a r t i c l e i n f o a b s t r a c t

Article history: Geologic secular trends are used to refine the timetable of supercontinent assembly, tenure, and breakup. The
Received 13 July 2010 analysis rests on what is meant by the term supercontinent, which here is defined broadly as a grouping of
Accepted 17 May 2011 formerly dispersed continents. To avoid the artificial pitfall of an all or nothing definition, quantitative
Available online 26 May 2011
measures of “supercontinentality” are presented: the number of continents, and the area of the largest
continent, which both can be gleaned from global paleogeographic maps for the Phanerozoic. For the secular
Keywords:
Supercontinent
trends approach to be viable in the deep past when the very existence of supercontinents is debatable and
Secular trends reconstructions are fraught with problems, it must first be calibrated in the Phanerozoic against the well
Time series constrained Pangea supercontinent cycle. The most informative geologic variables covering both the
Precambrian geology Phanerozoic and Precambrian are the abundances of passive margins and of detrital zircons. Both fluctuated
Passive margin with size of the largest continent during the Pangea supercontinent cycle and can be quantified back to the
Detrital zircon Neoarchean. The tenure of Pangea was a time represented in the rock record by few zircons and few passive
margins. Thus, previously documented minima in the abundance of detrital zircons (and orogenic granites)
during the Precambrian (Condie et al., 2009a, Gondwana Research 15, 228 242) now can be more confidently
interpreted as marking the tenures of supercontinents. The occurrences of carbonatites, granulites, eclogites,
and greenstone belt deformation events also appear to bear the imprint of Precambrian supercontinent
cyclicity. Together, these secular records are consistent with the following scenario. The Neoarchean
continental assemblies of Superia and Sclavia broke up at ca. 2300 and ca. 2090 Ma, respectively. Some of their
fragments collided to form Nuna by about 1750 Ma; Nuna then grew by lateral accretion of juvenile arcs
during the Mesoproterozoic, and was involved in a series of collisions at ca. 1000 Ma to form Rodinia. Rodinia
broke up in stages from ca. 1000 to ca. 520 Ma. Before Rodinia had completely come apart, some of its pieces
had already been reassembled in a new configuration, Gondwana, which was completed by 530 Ma.
Gondwana later collided with Laurentia, Baltica, and Siberia to form Pangea by about 300 Ma. Breakup of
Pangea began at about 180 Ma (Early Jurassic) and continues today. In the suggested scenario, no
supercontinent cycle in Earth history corresponded to the ideal, in which all the continents were gathered
together, then broke apart, then reassembled in a new configuration. Nuna and Gondwana ended their
tenures not by breakup but by collision and name change; Rodinia's assembly overlapped in time with its
disassembly; and Pangea spalled Tethyan microcontinents throughout much of its tenure. Many other secular
trends show a weak or uneven imprint of the supercontinent cycle, no imprint at all. Instead, these secular
trends together reveal aspects of the shifting background against which the supercontinents came and went,
making each cycle unique. Global heat production declined; plate tectonics sped up through the Proterozoic
and slowed down through the Phanerozoic; the atmosphere and oceans became oxidized; life emerged as a
major geochemical agent; some rock types went extinct or nearly so (BIF, massif type anorthosite, komatiite);
and other rock types came into existence or became common (blueschists, bioclastic limestone, coal).
Published by Elsevier B.V.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2. What constitutes a supercontinent? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3. Pangea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4. Secular trends that reflect the supercontinent cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.1. Area of the largest continent and number of continents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

☆ Bradley, D.C., 2011. Secular trends in the geologic record and the supercontinent cycle. Earth-Science Reviews ##, ##-##.
⁎ Tel.: + 1 907 786 7434; fax: +1 907 786 7401.
E-mail address: dbradley@usgs.gov.

0012-8252/$ – see front matter. Published by Elsevier B.V.

doi:10.1016/j.earscirev.2011.05.003
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 17

4.2. Abundance, start dates, and end dates of passive margins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

4.3. Granites and detrital zircons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.4. Isotopic composition of seawater strontium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.5. Deformation ages of greenstone belts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.6. Eclogite and granulite facies metamorphism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.7. Carbonatites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.8. Large igneous provinces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5. Proposed supercontinent timetable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.1. Vaalbara, Superia, and Sclavia (Bleeker, 2003) or Kenorland (Williams et al., 1991) . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2. Nuna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.3. Rodinia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.4. Gondwana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6. Supercontinent cycles and other secular variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.1. Global heat production and rates of plate tectonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.2. Oceanic crust and passive margin proxies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.3. Obducted ophiolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.4. High pressure, low temperature metamorphic rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.5. Mantle derived igneous rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.6. Massif type anorthosites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.7. Orogenic gold deposits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.8. Sedimentary rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.9. Sedimentary recycling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.10. Oxygenation of the atmosphere and oceans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.11. Glaciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6.12. Sea level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6.13. Mississippi Valley type lead zinc deposits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6.14. Other secular trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

1. Introduction Barley and Groves, 1992; Condie, 2005; Condie et al., 2009a;
Hawkesworth et al., 2010) and is well suited to study by this
Global paleogeography and plate interactions are reasonably well means because the opening and closing of ocean basins impact many
understood back to the assembly of Pangea in the late Paleozoic parts of the Earth system. In this paper, I examine the timing of past
(Fig. 1). Multiple lines of evidence go into these reconstructions supercontinent cycles using a new compilation of secular trends for
marine magnetic anomalies, passive margin matchups, geologic most of the geologic datasets for which this sort of information is
interpretation of orogenic belts, paleomagnetism, paleobiogeogra now available: age distributions of rocks and minerals, geochemical
phy of fossils, and distribution of climatically sensitive strata and trends, censuses of tectonic settings, numerical model results, and
the results are synthesized into a scenario that is consistent with the more. Many new global secular trends are published every year. The
rules of plate kinematics. Like genealogy, however, the plate most informative secular trends for present purposes are those that
reconstruction approach to Earth history gets harder and harder show fluctuations related to the assembly, tenure, and disassembly of
back though time, as each trail of evidence gets fainter. It would be Pangea. It will be shown that the tenure of Pangea was a time of few
pointless, for example, to attempt a global plate reconstruction at passive margins and low zircon abundance. These two variables can
4030 Ma, with only the Acasta Gneiss to go on. Precambrian be tracked deeper into the past to deduce, on uniformitarian grounds,
reconstructions are easiest when continents were clustered togeth the tenures of putative Precambrian supercontinents. A number of
er and there were fewer objects; but even these times of secular trends show this same irregular pulse, with maxima and
supercontinent tenure are a challenge. For example, Neoproterozoic minima at several hundred million years' spacing (Figs. 11 23). Other
reconstructions of Rodinia (Moores, 1991; Karlstrom et al., 2001; Li secular trends show little or no sign of it, however, but instead paint a
et al., 2008; Evans, 2009) show West Africa in very different changing backdrop against which the supercontinents have come and
positions. If it is that hard to decide where it goes, the assumption gone (Figs. 24 45).
that West Africa was part of a Rodinia supercontinent is itself open The terms “secular trend”, “time trend”, and “time series” are used
to question. here as synonyms for a set of ordered pairs (x, y) where y is a geologic
A complementary approach to Earth history is the analysis of variable and x is its age. The new compilation expands on previous
secular trends (e.g., Condie, 2005; Dewey, 2007; Reddy and Evans compendia of secular trends by Garrels and Mackenzie (1971), Meyer
2009; Condie et al., 2011; Goldfarb et al., 2010). Countless geologic (1981, 1988), Nance et al. (1986), Hallam (1992), Barley and Groves
variables can be tracked through time, even times when plate (1992), Groves et al. (2005), Condie (1997, 2005), and Veizer and
reconstructions are out of the question. This makes it possible to Mackenzie (2003). A few plots are new, either constructed from
sidestep the unknown specifics of plate paleogeography and instead data that were given only in tabular form in the original publications,
focus on evolution of the Earth system as a whole. Two precepts guide or from my own compilations. The plots are divided between the
this approach: (1) every secular trend can have only one correct main body of the paper (Figs. 3 45; Table 1) and a Supplementary
explanation (complicated though it might be); and (2) any viable data section (Figs. A1 A86; Table A1), available online. Tables 1 and
explanation for one secular trend must honor all the rest. A1 serve as figure captions. The Supplementary data section
The supercontinent cycle (Fig. 2) is manifested in a number of includes discussions of the data or model assumptions behind some
secular trends (e.g., Worsley et al., 1984, 1986; Nance et al., 1986; plots.
 Author's personal copy

18 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

A. Permian-Carboniferous boundary, 299 Ma B. Late Jurassic, 156 Ma

Fig. 1. Paleogeographic plate reconstructions from Stampfli and Borel (2002). (A) Permian–Carboniferous boundary, about 299 Ma. Pangea is in its final stages of assembly. Although
roughly 95% of the continental area was in Pangea, a series of microcontinents dotted the Paleotethyan ocean, in transit from Gondwana to Eurasia. (B) Late Jurassic, about 156 Ma.
The main Pangean landmass is shown in an early stage of breakup. On a constant-radius Earth, opening of the central Atlantic between Africa and North America (red arrows) must
have been accompanied by the subduction of an equivalent amount of seafloor, most likely along North America's Cordilleran margin.

All secular trends have been plotted or replotted with time as the supercontinent cycle is periodic (as suggested by Korenaga, 2006, among
x axis and the present on the left 1. Plots are presented at two time others) or stochastic. What is ultimately needed is a numerical index of
scales, either covering the Phanerozoic plus the very end of the “supercontinentality”, as discussed further in Section 4.1.
Neoproterozoic (550 Ma to present), or spanning Earth history in its A single round of the supercontinent cycle refers to the Earth as a
entirety (4560 Ma to present). Numerical age assignments use the whole and involves the breakup of one supercontinent and the
time scale of Gradstein and Ogg (2004), except for a few plots that subsequent assembly of a new one. A corollary of the inclusive
have an older time scale inextricably embedded in the construction. definition is the allowance that all the same pieces need not be parts
of both the first supercontinent and the second. Indeed, as noted by
Bleeker (2003), no supercontinent cycle in Earth history can be
2. What constitutes a supercontinent? proven to have conformed to the ideal cycle, with all of the continents
gathered into one, then breakup, and then reassembly of all of the
Multiple lines of evidence suggest that there have been times when pieces into some new configuration.
some formerly independent continents came together, and other times The terms “supercontinent cycle” and “Wilson cycle” are not
when larger continents fragmented into smaller ones. When is a synonymous. The Wilson cycle 2 refers to the opening and closing of a
grouping of continents big enough to earn the name supercontinent? single ocean basin. Despite a widespread misconception, there is no
Does this semantic distinction even matter? Many tectonicists use the stipulation in a Wilson cycle that the continent that rifts away is the
term supercontinent in the sense of Hoffman (1999): “a clustering of one that comes back; indeed this appears to be the rare case. A
nearly all the continents” or Rogers and Santosh (2003): “an assembly of supercontinent cycle is global and involves the aggregate effect of
all or nearly all the Earth's continental blocks”. I suggest that an “all or many Wilson cycles that are only roughly synchronous. The timing of
nearly all” definition (1) sets the bar higher than the rock record a Wilson cycle can be closely specified. The start date is when seafloor
requires, and (2) is impossible to rigorously apply in the Precambrian, spreading begins during the separation of two continents, and the end
when plate reconstructions are equivocal. Some researchers treat date is when the last of this seafloor is consumed during collision. The
Gondwana as one of the supercontinents (e.g., Condie 2005; Korenaga, timing of a supercontinent cycle as a whole can't be easily specified
2006) but it fails to meet the “all or nearly all” definition, because it because the component Wilson cycles aren't perfectly synchronized.
didn't include Siberia, Baltica, or Laurentia, it became Pangea. Bleeker Hypothetical scenarios for supercontinent evolution are shown in
(2003) coined the term supercraton for clusters of continents that would Fig. 2. The more straightforward scenario (Fig. 2A E) corresponds to
not meet the “all or nearly all” requirement. I prefer a more inclusive what Murphy and Nance (2003, 2007) referred to as extroversion: the
definition of supercontinent: a grouping of formerly dispersed continents. supercontinent turns “inside out” (Hoffman, 1991). Supercontinent
The difference between the “all or nearly all” and “inclusive” definitions #1 breaks up, and the pieces eventually regroup elsewhere in a
is more than just semantics, because it bears on whether or not the completely new configuration as supercontinent #2. In Fig. 2A C,
breakup of supercontinent #1 is enabled by subduction along two
1
This is now the universal convention in detrital zircon geochronology, the
2
subdiscipline that accounts for the vast majority of geologic time series now being The concept was first elaborated by Wilson (1968); the name was coined by
published. Dewey and Burke (1974).
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 19

Andean type margins, much as the opening of the Atlantic during Extroversion Introversion
Pangea's breakup was accompanied by Cordilleran subduction along
the Pacific rim during the Mesozoic (Collins, 2003). After the
supercontinent splits in two, each succeeding frame follows naturally
from the one before, eventually leading to formation of superconti
nent #2 on the other side of the globe. Supercont.
The more complex, hypothetical case known as introversion is #1 A Supercont.
#1 F
shown in Fig. 2F J. Supercontinent #1 breaks up, and its pieces move
apart for some distance but then reverse course and collide to form KEY:
supercontinent #2 in about the same place as before. During the continent
oceanic crust:
reversal, two subduction zones and a spreading center must die, and youngest
two new subduction zones and a spreading center must be born. Z Z
S1 S1 oldest
3. Pangea
P P
Pangea is the most recent and best understood supercontinent. Its R1 B R1
G
P P
basic configuration was appreciated a century ago by the early
proponents of continental drift, Taylor and Wegener. Pangea's late S2 S2
Paleozoic to early Mesozoic tenure (ca. 310 to 180 Ma) is well Z Z

constrained. Working back from the present, Pangea is reconstructed

simply by fitting the continents that border Atlantic type oceans: Z
P P
S1
Europe Greenland, Africa North America, India Antarctica, Australia dies
S1
Antarctica, and so on.
The Pangea fit also has to account for continents that migrated south
to north across Paleotethys and Neotethys during the late Paleozoic and R1 C Z
S3
S4 H
R2

Mesozoic (Fig. 1) (e.g., Sengör et al., 1988; Stampfli et al., 1991; Stampfli
and Borel, 2002). In the absence of marine magnetic anomalies, this part
S2 S2
of the puzzle relies on the art of deciphering orogenic belts. The Tethyan dies
continents are what keep Pangea from being an “ideal” supercontinent: P P
Z
it never contained all of the continental crust.
P
4. Secular trends that reflect the supercontinent cycle
S1
Z
4.1. Area of the largest continent and number of continents R2 D Z
S3
S4
I R2

S2 Z
Although differing in detail, all published Phanerozoic plate
reconstructions (e.g., Scotese, 1997; Stampfli and Borel, 2002) plainly
show that the continents were dispersed, then gathered together, and P
then dispersed once again. Two secular trends, derived for the present
study from the maps of Scotese (1997), serve as quantitative indices P
of “supercontinentality”.
In Fig. 3 10, the area of the largest single continent was tracked
through time by summing the areas of the main pieces, using values of E Z Z C R2
J
R2
C
continental areas from Cogley (1984). Today, there are five major dies dies
continental masses: South America, Africa, Australia, Antarctica, and the Supercont.
largest one, Eurasia plus North America. The latter two are joined across #2
P
the logjam of Phanerozoic accreted terranes in Alaska and the Russian
Far East, and I count them as one (even though they are kinematically
separate plates) because their mutual boundary is long, diffuse, and
Fig. 2. Schematic cross sections through the Earth showing two idealized modes of
relative motions are extremely slow. I count Arabia and India as part of supercontinent cyclity, showing many of the same concepts as a figure of Silver and Behn
Eurasia on the grounds that the Himalayan and Zagros sutures have long (2008). Passive margins related to each cycle are indicated by a “P” and zircon-forming
since closed, and I count Greenland with North America because the two environments by a “Z”. R1 and R2 identify spreading ridges; S1 to S4 identify subduction
zones; C indicates collisional orogens. In order to emphasize only those plate motions directly
never completely separated and are no longer in relative motion. North
involved in the supercontinent cycle, preexisting plate boundaries are not shown in the first
America and South America are considered separate for this analysis, frame of each sequence; the simplest scenarios would commence with subduction zones
because even though they are presently linked by a narrow arc, the already in existence, as indicated by the dashed slabs in A and F. (A-E) Extroversion. Breakup
connection appears to be ephemeral; if they appeared in the same of supercontinent #1 is enabled by subduction along two Andean-type margins (S1 and S2).
relative positions in an older reconstruction, they would certainly be The two halves of supercontinent #1 eventually collide (C) on the other side of the world to
form supercontinent #2. (F-J) Introversion. At first, the sequence is the same as for
treated as separate. Working back through time, the same kinds of
extroversion, but in order to return to where they began, the two diverging continents
decisions are needed for each time slice but with less information; the somehow must reverse course. This implies that the two original subduction zones (S1 and
resulting time trend (Fig. 3 10) is subjective in detail, though certainly S2) must die, that the seafloor that formed at spreading center R1 must now be consumed at
correct in broad form. one or more new subduction zones (e.g., S3 and S4), and that one new spreading ridge (R2)
must appear.
A related way of tracking supercontinent assembly and disassembly
in the Phanerozoic is by the number of independent continents (Fig. 4).
The trouble is that every continent counts as one regardless of size, and
the size cutoff for whether or not to count a microcontinent is arbitrary.
Two versions of this time trend are shown, one from Worsley et al.
 Author's personal copy

20 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

Ceno- Mesozoic Paleozoic

Ceno- Mesozoic Paleozoic
zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed
zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed.

Relative sea level (meters x 100)

100
200
7

Percent of continental area

3 6 Relative sea level
75 5 (Hallam, 1984)
Area (106 km sq)

150
4

70 3
100
2
Area of largest continent 1
50 25
(this study; from Scotese, 1997)
0

0 0 -1
0 100 200 300 400 500 0 100 200 300 400 500

60

6 This study 4 8
Number of continents

40
Worsley et al. (1984)

Area (million km)

Age distribution of oceanic crust
4 30
(Sclater et al., 1980)

20
2
10
Number of continents
0 0
0 100 200 300 400 500 0 100 200 300 400 500

80
Passive margin abundance
40 scale change
(Bradley, 2008) 5 25 9
Age distribution
20
Margins per bin

20 of ophiolites
modern
margins
(Dilek, 2003)
Number

15 15

10 10
ancient
margins
5 5

0 0
0 100 200 300 400 500 0 100 200 300 400 500
Preserved sedimentary mass (%)

.710 Strontium isotopic 6 6

10
composition of seawater on oceanic crust
Sedimentary mass
(Shields & Veizer, 2002) (Veizer & Mackenzie, 2003)
Sr/86Sr

4
.708 on continental crust
on passive margins
87

.706 0
0 100 200 300 400 500 0 100 200 300 400 500
Age (Ma) Age (Ma)

Figs. 3-10. Secular trends for 0 to 550 Ma. (Table 1 takes the place of these captions).

(1984) and the other derived for the present study from the re cycle controls the global abundance of passive margins, because passive
constructions of Scotese (1997). Whereas the time trends differ in detail, margins are created during supercontinent breakup and are destroyed
the tenure of Pangea corresponds to a broad minimum in both plots. during supercontinent assembly (Fig. 2). The present day passive
Precambrian plate reconstructions are not nearly as robust as margins those flanking extant Atlantic type oceans total about
Phanerozoic ones, so quantitative measures of “supercontinentality” 100,000 km in length and range in age from a few million years (Red
area of the largest continent and number of continents are fraught Sea) to about 180 million years (east coast of Africa). Analysis of ancient
with uncertainty. Proxies are needed. passive margins is not as straightforward because they no longer face an
ocean, but instead are preserved in orogenic belts (Bradley, 2008).
4.2. Abundance, start dates, and end dates of passive margins The Phanerozoic age distribution of passive margins (Fig. 46) tracks
the assembly, tenure, and disassembly of Pangea. The tenure of Pangea
Passive margins form along the matching edges of continents that coincided with a low at 300 to 275 Ma. The increase in passive margins
rift and are then carried apart by seafloor spreading. The supercontinent since then tracks Pangea's breakup; the sharp decline from ca. 500 to
 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
80 250 14
282
40 ++ modern margins 11 15 19
scale change Passive-margin Detrital zircons from modern river sands 12 Collision ages of
+ quality ranking D 200
20 abundance quality ranking C (Campbell and Allen, 2008) greenstone belts
quality ranking B
10
+ (Bradley, 2008) 165 (Condie, 1994)
quality ranking A 150
15 8
+ 447 1035
1871

Number
6

Number
100
10 ++
+ 2706

per age bin

4
5 50

Number of margins
+ 2
n = 5246
0 + 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

30 174
20 12 1600 16 20
Passive-margin start dates
10 Detrital zircons from sandstones 20 Age distribution of eclogites
(Bradley, 2008)
and metasandstones (Brown 2007)
scale change 1200
(this study) 15

6
800 432 1824 10

Number
Number

Number
4 594
1047
400 1435 5
2 2697
2500
367 529 n = >26,000
1371
0 0 853 1565 0
2347
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

.710 20
8 n = 5059
13 17 Age distribution of granulites 21
.708 and ultra-high temperature
15
6 metamorphic rocks
Passive-margin end dates .706 (Brown, 2007)
4 (Bradley, 2008) 10
seawate r granulites
.704 evolution

Number
87Sr/86Sr
Number

ma ntle
c ontr ib 5 UHT
2 ution (Shields & Veizer,
.702
Strontium isotopic
D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33
Author's personal copy

2002)
composition of seawater
0 .700 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

700 100 50
U-Pb ages of granites 2700 14 18 22
600 Age distribution of carbonatites
(Condie et al., 2009a) 80 Continental component 40
(Woolley & Kjarsgaard, 2008)
500 1893 of seawater 87Sr/86Sr
60 (Shields, 2007) 30
400
1707 2501
300
Number

Number
20

Percent
1654 2104
40
569
200 1376
1046
20 10
100

0 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

Age (Ma) Age (Ma) Age (Ma)

21
 22
Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
5
600
16 Age distribution of 27 31
Large Igneous
23 Lifespans of Pressures in
500 passive margins 4
Provinces metamorphic rocks
12 (Bradley, 2008) (Brown, 2008)
(Prokoph et al., 2004) 400
3
UHP
8 300
2 lawsonite
E-HPGM

Number
200

Lifespan (m.y.)
granulite
4

Pressure (GPa)
1
100 G-UHTM

0 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

8 50
600
7 Global heat production Age of oldest 28 32
24 500 Atlantic-type oceanic c rust 40
6 (Pollack, 1997)
(this study)
5 400
30 Geothermal
4 300 N gradients in
S-S
,000 R 20 metamorphic
3 = 10

Age (m.y.)
K/U 200 rocks

Gradient °C/km
2 00 1500
= 65,0 K /U = (Brown, 2008)
K/U 10
1 100 P

0 0 0

Heat production (past/present)

0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

1800 75 10 2747
komatiite 25 Age distribution of ophiolites 29 Age distribution 33
basalt 60 (Moores, 2002; Dilek, 2003) 8
of komatiites
1600 (Isley and Abbott, 1999)
45 6

T°C
3406
Number

30 4

Number
1400
D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

Potential temperature
Author's personal copy

of the mantle 15 2
(Herzberg et al., 2010)
1200 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

10
12 MORB
30

l
26 20 Age distribution of blueschists Neodymium isotopic 34

na
10

io
Global plate velocity compositions of mantle-derived rocks

nt
and ultra-high pressure

ve
n (Korenaga, 2006) metamorphic rocks 8 d
(Bennett, 2003)
co 15 ep
(Brown, 2007) let
e dm
5 6 an
tle
εNd

10

U (cm/yr)
4

Number
γ 0 = 0.30 Ultra-high pressure
5 2
Lawsonite-blueschist
γ 0 = 0.15 & lawsonite-eclogite primitive mantle
0
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)
 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
4.0 300
primitive mantle
3.8 Density of subcontinental
35 4 39 43
250
lithospheric mantle shale
4.6 Molybdenum
(Poudjom-Djomani et al., 2001) 3 200
3.4
concentrations in shale
sandstone (Scott et al., 2008)
150
3.2 2

Mo (ppm)
mafic lower crust 100

Density (g/cm3)
3.0 Neodymium isotopic
intermediate lower crust 1 compositions of sedimentary rocks
50

Mantle sep. age (Ga)

2.8 (Vervoort et al., 1999)
felsic lower crust
2.6 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

36 4 40 44
6 Age distribution of
massif anorthosites Glaciations
3 (Hoffman, 2009)
(Ashwal, 1993)
4
2

Number
Igneous age
versus depositional age blue line = global
2 green line = regional

Zircon age (Ga)

1 of detrital zircons
(this study)
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

500 12

37 Mass-independent 40 45
Age distribution of 41
400 fractionation of 35
orogenic gold deposits 8
Goldfarb et al., 2001a) sulfur isotopes 30 Mississippi Valley-type
300 (Domagal-Goldman lead-zinc deposits
25
et al., 2008)

∆S
4 (Leach et al , 2010)
20

33
200
15
Pb+Zn (Mt)

0 10
100
D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33
Author's personal copy

Gold (million ounces)

0 -4 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma)

Abundance of iron formation

38 (Bekker et al., 2010)
42
106
20 Age distribution of sedimentary rocks Rapitan
105
(Ronov et al., 1991) type
104
Superior
103
type
10 102
101

Iron formation (Gt)

100

Sedimentary mass (1020 g/m.y.

0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma)

Figs. 11-45. Secular trends for 0 to 4560 Ma. (Table 1 takes the place of these captions).
23
 Author's personal copy

24 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

Table 1
Summaries of time-series plots in Figs. 3-43.

Fig. no. Variable Time span Reference and comments

(Ma)

3 Area of largest continent 0-550 This study, derived from Scotese (1997)
4 Number of continents 0-550 This study and Worsley et al. (1984)
5 Passive margin abundance 0-550 Bradley (2008)
6 Strontium isotopic composition of seawater 0-550 Shields and Veizer (2002)
7 Relative sea level 0-550 Hallam (1992); similar to the curve by Haq et al. (1987) and
Haq and Schutter (2008) but constructed on different principles
8 Age distribution of oceanic crust 0-550 Sclater et al. (1980)
9 Age distribution of ophiolites 0-550 Dilek (2003)
10 Sedimentary mass 0-550 Veizer and Mackenzie (2003) using a plot from Hay
11 Passive-margin abundance 0-4560 Bradley (2008)
12 Passive-margin start dates 0-4560 Bradley (2008)
13 Passive-margin end dates 0-4560 Bradley (2008)
14 U–Pb ages of granites 0-4560 Condie et al. (2009a)
15 Detrital zircons from modern river sands 0-4560 Campbell and Allen (2008); a similar plot was published by Rino et al. (2008)
16 Detrital zircons from sandstones and metasandstones 0-4560 This study
17 Strontium isotopic composition of seawater 0-4560 Shields and Veizer, 2002
18 Continental component of seawater 87Sr/86Sr 0-4560 Shields (2007)
19 Collision ages of greenstone belts 0-4560 Condie (1994)
20 Age distribution of eclogites 0-4560 Brown (2007); medium T eclogite-high P granulite
21 Age distribution of granulites and ultra-high temperature metamorphic rocks 0-4560 Brown (2007)
22 Age distribution of carbonatites 0-4560 Woolley and Kjarsgaard (2008)
23 Age distribution of Large Igneous Provinces 0-4560 Prokoph et al. (2004)
24 Global heat production 0-4560 Pollack (1997). Red curve shows favored mix of radioactive fuels.
25 Potential temperature of the mantle 0-4560 Herzberg et al. (2010). Curves represent model calculations based
on different Urey ratios.
26 Global plate velocity 0-4560 Korenaga (2006). Black and red curves represent different Urey ratios.
27 Lifespans of passive margins 0-4560 Bradley (2008)
28 Age of oldest Atlantic-type oceanic crust 0-4560 Bradley (2008) and this study. Dashed segment is poorly constrained.
29 Age distribution of ophiolites 0-4560 Dilek (2003); Moores (2002)
30 Age distribution of blueschists and ultra-high pressure metamorphic rocks 0-4560 Brown (2007)
31 Pressures in metamorphic rocks 0-4560 Brown (2008)
32 Geothermal gradients in metamorphic rocks 0-4560 Brown (2008)
33 Age distribution of komatiites 0-4560 Pollack (1997)
34 Neodymium isotopic compositions of mantle-derived rocks 0-4560 Bennett (2003)
35 Density of subcontinental lithospheric mantle 0-4560 Poudjom Djomani et al. (2001)
36 Age distribution of massif anorthosites 0-4560 Ashwal (1993)
37 Age distribution of orogenic gold deposits 0-4560 Goldfarb et al. (2001a)
38 Age distribution of sedimentary rocks 0-4560 Ronov et al. (1991)
39 Neodymium isotopic compositions of sedimentary rocks 0-4560 Vervoort et al. (1999)
40 Igneous age versus depositional age of detrital zircons 0-4560 This study
41 Mass-independent fractionation of sulfur isotopes 0-4560 Domagal-Goldman et al. (2008)
42 Abundance of iron formation 0-4560 Bekker et al. (2010)
43 Molybdenum concentrations in shale 0-4560 Scott et al. (2008)
44 Glaciations 0-4560 Hoffman (2009) and unpublished
45 Mississippi Valley-type lead-zinc deposits 0-4560 Leach et al. (2010)

ca. 350 Ma coincides with Pangea's assembly. Thus the Phanerozoic 2008). Maxima in this plot at 2050 2000, 650 600, 550 500, and
portion of the passive margin age distribution forms a clear pattern with 150 100 Ma represent times of synchronized continental breakup.
a straightforward connection to the Pangea supercontinent cycle. For the vast majority of passive margins that met their end by
Deeper in the past (Fig. 11), the abundance of passive margins colliding with a convergent margin, the end date is taken as the time
increased between ca. 2750 and 2500, between ca. 2300 and 2050, when the continent ocean boundary entered the subduction zone at
and between ca. 1050 and 650 Ma, and it declined from ca. 2500 to the start of arc passive margin collision. Maxima are at 1850 1800,
2400 and ca. 1850 to 1750; these times bracket lows at ca. 2400 2300 600 550, 450 300, and 100 50 Ma. Each start date and each end date
and ca. 1750 1650 Ma (Bradley, 2008). There is no evidence that the represents a plate reorganization involving the formation or death of a
uneven age distribution is primarily due to uneven preservation. divergent plate boundary and the formation or death of a convergent
Instead, applying the patterns from the Phanerozoic, the fluctuations plate boundary (cf. Fig. 2). The broad age distributions in Figs. 12 and
are inferred to mark times of supercontinent disassembly, assembly, 13 are similar but slightly offset, as would be expected.
and tenure, respectively. Superimposed on the various fluctuations,
the passive margin age distribution is dominated by younger margins, 4.3. Granites and detrital zircons
which I attribute to a combination of destruction of older margins,
difficulties in recognizing and dating the vestiges of old margins in Condie et al. (2009a) showed that granites and detrital zircons that
ancient orogens, and different counting conventions between modern were ultimately derived from them have remarkably similar, strongly
and ancient margins (Bradley, 2008). The utility of passive margin episodic age distributions. The fluctuations have been interpreted in
abundance as a proxy for supercontinent cycles hinges on the terms of Precambrian supercontinent cycles (e.g., Condie et al., 2009a).
accuracy and thoroughness of the census, as discussed further in the Both granites and detrital zircons have advantages and shortcomings for
Supplementary data section. this line of inquiry. U Pb zircon ages of granites specifically, those
The start dates and end dates of these passive margins are shown classed as orogenic granites (Condie et al., 2009a) provide the more
in Figs. 12 and 13. The start date of a passive margin is taken as the direct record. The recent global census of igneous U Pb ages by Condie
onset of seafloor spreading (i.e., the rift drift transition) (Bradley, et al. (2009a) is shown in Fig. 14. The robust, Precambrian part of this
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 25

Fig. 46. (A) Published assessments of the tenures of various supercontinents according to the identified authors. (B) Age distributions (in black) of variables that have bear on the
tenures of supercontinents, from sources cited in Figs. 11, 14, 15, 16, and 20. For each plot, the blue, green, lavendar, orange, and pink swaths indicate tenures of supercontinents as
inferred from minima in those data alone. Dimmer and brighter colors represent more and less inclusive interpretations, respectively. The colored swaths agree in general but differ
in many details. (C) Proposed tenures of supercontinents based on the present study, combining information from Phanerozoic plate reconstructions, passive-margin age
distributions, and zircon age distributions.

secular record shows probability maxima in granite abundance at 2700, and leaves out, or barely accounts for, enormous magmatic provinces
1893, and 569 Ma, as well as a number of subsidiary peaks. such as the Okhotsk Chukotsk belt, the Patagonia batholith, and the
This approach does have a number problems: (1) global data Sierra batholith. Even knowing when to look for it, the tenure of
coverage is uneven, being thinnest in remote areas; (2) countless Pangea cannot be discerned in this age distribution. Accordingly, there
plutons have been lost from the geologic record, either buried beneath is direct no way to calibrate Fig. 14 against the only unequivocal part
sedimentary cover, or utterly gone like the Hadean source plutons of of the supercontinent cycle.
the Jack Hills detrital zircons; (3) classification of some granites as Detrital zircon data take care of some of the shortcomings of the
“orogenic” or not can be equivocal; and (4) 40Ar/ 39 ages were not granite data. Two general approaches have been used in global
compiled, resulting in an undercount of Mesozoic and Cenozoic compilations of detrital zircons. Rino et al. (2008) and Campbell and
igneous rocks. The Phanerozoic part of the census is meager indeed, Allen (2008) each dated thousands of detrital zircons from modern sands
wrongly giving the impression that zircon production has declined to of major rivers, the idea being that these rivers collect a representative
zero since the Jurassic. The compilation includes no modern volcanoes sampling of zircon from their drainages. Fig. 15 shows Campbell and
 Author's personal copy

26 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

Allen's (2008) histogram based on 40 rivers. Maxima are seen at water 87Sr/ 86Sr is plotted for the Phanerozoic in Fig. 6 and for the
2706, 1871, 1035, 447, 282, and 165 Ma and minima at 2565, 2247, 1576, entire span of Earth history in Fig. 17. These isotopic ratios were
1294, 929, 671, 424, and 2233 Ma (Fig. 15). This compilation, and a measured from over five thousand marine calcite samples shells
similar one by Rino et al. (2008) (Fig. A21) share one shortcoming: the from the Phanerozoic and limestones from the Precambrian (Shields
world's biggest rivers mainly traverse ancient cratons. Young orogenic and Veizer, 2002). The lowest values of a given age are taken to most
belts are mostly drained by smaller rivers and there are no rivers of any closely track the composition of seawater. The seawater curve
size on many island arc volcanoes. Consequently, the paucity of detrital represents the ever shifting balance between sources of primitive
zircons younger than about 150 Ma is at least partly an artifact. mantle strontium (low 87Sr/ 86Sr that systematically rose through
Detrital zircons from sandstones and metasandstones provide time), which mostly enters seawater via hydothermal circulation at
another approach to global zircon distributions. The recent surge in mid oceans ridges, and evolved continental strontium (high but
the use of detrital zircon geochronology in studies of provenance and regionally variable 87Sr/ 86Sr), which is carried to the sea by rivers and
regional tectonics has already yielded at least a million individual by fresh and saline groundwaters (Veizer and Mackenzie, 2003, and
zircon dates. Fig. 16 shows a new compilation of N26,000 zircon ages. references therein). Because 87Sr/ 86Sr is a ratio, uneven preservation
Of the N350 sandstone and metasandstone samples in this database, is less of a problem than for variables that require counting such as
21 are Archean, 69 are Proterozoic, and 122 are Cambrian through passive margin abundance.
Triassic, and 140 are Jurassic or younger. Only this last group of In the Phanerozoic, the 87Sr/ 86Sr data show a sawtooth descent
samples has the potential to bear on zircon abundance during the from ca. 500 to ca. 260 Ma, a double low at ca. 260 and ca. 160 Ma, and
Pangea supercontinent cycle. The overall age distribution shows a long term rise rise from ca. 160 Ma to the present. This first order
maxima at 2729, 1824 1793, 1435, 1047, 594, 432, 141, and 76 Ma pattern broadly tracks the Pangea supercontinent cycle, with the
and minima at 2374, 1565, 1371, 853, 529, and 367 Ma. shorter term fluctuations corresponding, perhaps, to individual
The fact that three independent zircon datasets (Figs. 14, 15, and Wilson Cycles. The maxima are most likely the result of collisional
16) show comparable age distributions means that this cannot be an orogenesis, which results in erosion of large volumes of high 87Sr/ 86Sr
artifact of uneven sampling (Condie et al., 2009a). The possibility rock (Richter et al., 1992) and the death of a spreading center,
remains, however, that it is a different kind of artifact one due to eliminating an erstwhile source of low 87Sr/ 86Sr.
uneven preservation. Hawkesworth et al. (2010) took the latter For most of the Precambrian, age resolution is not as good, and the
position, suggesting that zircons generated during supercontinent fine texture seen in the Phanerozoic 87Sr/86Sr seawater curve cannot yet
forming collisions are well situated to be preserved, whereas zircons be discerned (Veizer and Mackenzie, 2003) (Fig. 17). Nonetheless, first
along intraoceanic arcs are less so. This is an important problem. I order fluctuations are recorded. Fig. 18, which is derived from the same
submit that uneven preservation cannot readily account for the broad raw data, isolates the continental contribution to seawater 87Sr/86Sr by
similarities between the abundances of passive margins and detrital correcting for the mantle and carbonate weathering components
zircons (cf. Figs. 11 and 16), or fluctuations in seawater 87Sr/ 86Sr (Shields, 2007). The seawater 87Sr/86Sr curve first deviated from the
(Figs. 17 and 18), as discussed in Section 4.4. mantle evolution line at about 2800 Ma. A maximum in Fig. 18 at
Adapting the arguments of Silver and Behn (2008), I postulate that roughly 2000 to 1800 Ma coincides with the assembly of Nuna, and a
the first order fluctuations in zircon abundance can be related to the minimum at roughly 1100 to 800 Ma overlaps with the nominal tenure
supercontinent cycle as follows. (1) Most zircons are generated during of Rodinia. The six dashed sections of the curves in Figs. 17 and 18 were
plate convergence, either subduction or collision. (2) Supercontinent interpolated across times of sparse or nonexistent data. Judging from the
assembly involves subduction followed by continent continent colli Phanerozoic, low values of seawater 87Sr/ 86Sr would be expected at ca.
sion (Fig. 2). (3) The formation of a supercontinent extinguishes the 2300 Ma and ca. 850 Ma, and a high would be expected at ca. 1050 Ma.
convergent plate boundaries along which it formed. It follows that once New data will be needed to test these predictions.
a supercontinent has formed, global zircon production zircon should
decline sharply. The only way around this is if, upon the death of a 4.5. Deformation ages of greenstone belts
collisional plate boundary and without delay, the equivalent amount of
convergence begins to be taken up elsewhere on new or pre existing Greenstone belts are deformed accumulations of supracrustal rocks
subduction zones. Silver and Behn (2008) pointed out that subduction containing at least some volcanic component (Condie, 1994). The term
initiation has not accompanied the Himalayan collision, even though 50 is a catch all and includes rocks inferred to have formed in ocean floor,
million years have elapsed since collision started. (4) Hence, the tenure intra oceanic arcs, continental arcs, forearc, intra arc and back arc
of a supercontinent should be marked by a dearth of zircons that lasts basins, and oceanic plateau settings especially in the Precambrian.
as long as the supercontinent itself. (5) Breakup should coincide with Few Phanerozoic oceanic sequences have been categorized simply as
a rise in subduction related magmatism (and by inference, an increase greenstone belts. For both of these reasons, the age distribution of
in zircon production), because new plate divergence in one place greenstone belts (Fig. A22, from Condie, 1994) is of uncertain
on Earth must be matched by plate convergence elsewhere. significance. However, by combining rocks of many tectonics settings,
If above reasoning is correct, an implication is that zircon abundance deformation ages of greenstone belts (Fig. 19, from Condie, 1994)
can potentially serve as a global plate speedometer. What is needed is a together provide a global record of convergent tectonism in the
large suite of carefully chosen samples from various tectonic settings, Precambrian. Orogenies involving greenstone belts were most
from all parts of the world, and from each of the past 30 or more geons. abundant at ca. 2700, 1850, 1050, and 600 Ma; the Precambrian age
As a first approximation from the less carefully selected datasets in distribution is broadly similar to that of granites and detrital zircons
Figs. 14 16, it would appear that at ca. 2300 Ma, global subduction flux and a general connection with the supercontinent cycle is evident.
(Fs of Silver and Behn, 2008) was 10 20% as fast as during the peaks of
zircon abundance at ca. 1800 or 2700 Ma. 4.6. Eclogite and granulite facies metamorphism

4.4. Isotopic composition of seawater strontium The distribution of extreme metamorphic conditions through time
suggests links with the supercontinent cycle (Brown, 2007, 2008).
The global 87Sr/ 86Sr ratio has increased through time owing to the Exhumed eclogite facies metamorphic belts reached their peak abun
inexorable decay of the world's initial allotment of 87Rb to 87Sr. dance at 400 300 Ma, synchronous with the terminal collisions that
Rubidium, meanwhile, has been concentrated in the continents, and formed Pangea; older peaks, based on fewer instances, are seen at
continental rocks accordingly have elevated 87Sr/ 86Sr values. Sea ca. 2500 2400, 1900 1800, and 700 600 Ma (Fig. 20). Exhumed high
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 27

temperature (granulite facies) and ultra high temperature metamor 5.1. Vaalbara, Superia, and Sclavia (Bleeker, 2003) or Kenorland
phic belts reached their peak abundance at ca. 600 500 Ma with minor (Williams et al., 1991)
peaks at 2800 2500, 2100 1800, and 1100 1000 (Fig. 21). Given the
paucity of exposed high temperature metamorphic rocks younger than The earliest continental assemblies are quite conjectural. Williams
ca. 500 Ma, the Pangea cycle is of no help in relating this style of et al. (1991) postulated a single Neoarchean supercontinent, Kenor
metamorphism to supercontinents. In both plots, however, the land, that was assembled by ca. 2500 Ma. Later, Bleeker (2003)
Precambrian maxima broadly match those in the age distributions of identified three smaller continental groupings based on common
granites and detrital zircons (Figs. 14 16) and broadly correlate with the themes among the world's 35 known Archean crustal blocks:
supercontinent cycle (Brown, 2007). High pressure metamorphic rocks Vaalbara, Superia, and Sclavia. Vaalbara, the oldest, includes the
and metamorphic conditions are discussed in Section 6.4. Kaapvaal and Pilbara cratons and was proposed by Bleeker (2003) to
have existed from ca. 3470 to ca. 2700 Ma. Superia includes Canada's
Superior craton plus other continents that rifted from it during the
4.7. Carbonatites Paleoproterozoic (e.g., Karelia, Hearn); Bleeker (2003) gave its age
range as ca. 2700 to 2450 Ma. Sclavia includes Canada's Slave craton
The age distribution of carbonatites (Fig. 22, from Woolley and and various other cratons that rifted away from it during the
Kjarsgaard, 2008) is sufficiently similar to those of granites and detrital Paleoproterozoic (e.g., Dharwar, Zimbabwe); Bleeker (2003) sug
zircons that a general connection with the supercontinent cycle appears gested an age range from ca. 2600 to 2200 Ma. Based on passive
likely. Binned in 50 m.y. increments, peak abundances are at 50 0 and margin ages, Bradley (2008) suggested minor modifications to
150 100 Ma, and the most prominent minor peaks are at 250 200, 400 Bleeker's timing: Vaalbara from ca. 3470 to ca. 2685 Ma, Superia
300, 600 500, 1050 1000, 1900 1850, 2100 2000, and 2700 2650 Ma. from ca. 2700 to ca. 2300 Ma, and Sclavia from ca. 2600 to ca. 2090 Ma.
Carbonatites have not been reported from before ca. 2800, or between Bleeker (2003) coined the term supercraton for these continental
2400 and 2150, 1550 and 1450, or 1300 and 1250 Ma. Carbonatites are groupings because none of the three were considered large enough to
widely interpreted as being related to extensional tectonic regimes, and meet the “all or nearly all” definition of supercontinent. A key
specifically to continental breakup (e.g., Burke et al., 2003; Rukhlov and question is whether the assembly, tenure, and disassembly of these
Bell, 2010). In this regard, aspects of the age distribution are problematic. three continental masses would have caused global plate reorganiza
The dominance of Triassic and younger carbonatites is probably tions. In each case the answer is yes. The breakup of Superia, for
exaggerated by attrition of older carbonatites, an effect discussed in example, involved the initiation of at least four new divergent plate
Section 6.2. The minor peak at 250 200 Ma lies squarely during Pangea's boundaries surrounding the Superior craton, totaling thousands of
tenure but the peak at 400 300 coincides with Pangea's assembly. The kilometers in length. In order to accommodate these ridges, suitably
Precambrian maxima and minima are essentially the same as those in oriented subduction zones must either have formed or, if already in
the orogenic granite or detrital zircon age distributions (Figs. 14 16). existence, must have begun consuming seafloor at a faster rate.
Except for the high at 2100 2000 Ma, the Precambrian maxima are The abundances of orogenic granites and detrital zircons show
times dominated by continental assembly, not breakup. huge fluctuations during the span between the putative assembly of
Vaalbara and the breakup of Sclavia, ca. 3470 to 2090 Ma. The dearth
of granites between 2400 and 2200 Ma and of detrital zircons
4.8. Large igneous provinces between 2400 and 2300 Ma coincides the simultaneous tenures of
Superia and Sclavia. (An equivalent low during the putative tenure of
Large Igneous Provinces (“LIPs”) are outpourings of mafic volcanic Vaalbara is not in evidence.) The remarkable maxima in granite and
rocks and cogenetic intrusive rocks, commonly regarded as having detrital zircon abundances at ca. 2700 to 2600 Ma coincide with the
formed above mantle plumes. Young LIPs include continental flood proposed assembly times of both Superia and Sclavia. If, alternatively,
basalt provinces and oceanic plateaus. Ancient LIPs are recognized Vaalbara, Superia, and Sclavia occupied far removed portions of a
from flood basalts, layered mafic ultramafic intrusions, dike swarms, single Kenorland supercontinent (Bleeker, 2003, his Fig. 5a), the
and, more controversially, komatiites (see Section 6.5). The distribu zircon record (Figs. 14, 15, and 16) would suggest staged assembly
tion of LIPS through time is plotted in Fig. 23 from Prokoph et al. from ca. 2750 to 2450 Ma followed by a dead time of supercontinent
(2004), which was a minor update of the comprehensive synthesis by tenure lasting to ca. 2200 Ma (cf. Condie et al., 2009b).
Ernst and Buchan (2001). The Phanerozoic is dominated by a broad
peak at 200 0 Ma. This is likely magnified by the inclusion of the
many LIPs in the present ocean basins; most of these will end up being 5.2. Nuna
subducted. Before 180 Ma, the LIP record is almost entirely continen
tal and thus is not directly comparable to the younger record. The age Paleoproterozoic breakup of Sclavia and Superia (or alternatively,
distribution shows maxima at 2800 2700, 2200 2100, 1800 of Kenorland) led to a time of dispersed continents, as suggested by a
1700 Ma, and 1300 1200 Ma, and minima at 2400 2300, 1600 peak in the abundance of passive margins between ca. 2050 and ca.
1500, 900 800, and 500 300 Ma. These four minima coincide with 1900 Ma (Fig. 11). Each margin was eventually involved in a collision,
the tenures of supercontinents as deduced from the zircon and creating a new continental grouping and leaving no passive margins
passive margin records. at all from ca. 1740 to ca. 1600 Ma. This late Paleoproterozoic
continental grouping was called Nuna by Hoffman (1997) and
Columbia by Rogers and Santosh (2002). The case that all the world's
5. Proposed supercontinent timetable continents were gathered into a single grouping is not compelling
but under my definition of supercontinent, Nuna would qualify
The tenures of supercontinents are shown from published regardless. Various proposed configurations are quite different (cf.
literature in Fig. 46A. Fig. 46B shows the principal individual secular Hoffman, 1997, Rogers and Santosh, 2003, and Zhao et al. 2002).
records that bear on the timing of supercontinents. A proposed Hoffman (1997) suggested that Laurentia, Greenland, and Baltica had
supercontinent timetable is shown in Fig. 46C, modified from Bradley come together by 1800 Ma. According to Rogers and Santosh (2003),
(2008) subject to the age constraints provided by the time trends Columbia came together at ca. 1800 Ma and broke apart ca. 1500 Ma.
shown. The present study does not bear on the details of particular According to Zhao et al. (2002), it came together (but in a very
supercontinent reconstructions. different configuration) by a series of collisions from 2100 to 1800 Ma,
 Author's personal copy

28 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

grew by subduction accretion until ca. 1300 Ma, and then broke up, to tites (Figs. 19 22). Knowing to look for it, this age pattern can be dimly
reform soon thereafter as Rodinia. made out in the orogenic granite plot (Fig. 14). On the other hand,
The granite and detrital zircon records (Figs. 14 16) are consistent passive margin and detrital zircon abundances, which were in synch
with the late Paleoproterozoic assembly of a supercontinent. The during the assembly and early tenure of Nuna, fall out of phase during
detrital zircon maximum at 1824 Ma (Fig. 16) can be attributed to the putative assembly of Rodinia. Only two passive margins are known
Nuna forming collisions, and the ensuing decline in zircon abundance from ca. 1200 to 1050 Ma, when their number started gradually rise,
to a minimum at 1565 Ma can be attributed to the death of a number of eventually reaching 24 by 600 Ma. Considering the detrital zircon and
convergent plate boundaries when Nuna came together. passive margin records together, it would appear that orogenies during
The passive margin record is also consistent with the late the assembly of Rodinia did not involve passive margins (Bradley,
Paleoproterozoic assembly of a supercontinent. An abundance of 2008). During several hundred million years leading up to the Grenville
passive margins between ca. 1850 and 2050 Ma was followed by a orogeny, eastern Laurentia was the site of a long lived convergent
precipitous drop between ca. 1850 and 1750 (Fig. 11). This would margin (Hoffman, 1991; Whitmeyer and Karlstrom, 2007); when the
correspond to the assembly of Nuna and is consistent with the various last seafloor was consumed along this orogen, the global abundance of
published interpretations cited above. On the other hand, the passive passive margins is not known to have changed. I submit that the early
margin record (Bradley, 2008) provides no independent confirmation Neoproterozoic minimum in detrital zircon abundance at ca. 1000 to
of the proposal that subsequently, during the Mesoproterozoic, a 850 Ma best represents Rodinia's tenure. It appears, though, that one
supercontinent broke up and then re formed into Rodinia (cf. Rogers part of Rodinia was already being disassembled (i.e., along Baltica's
and Santosh, 2003 and Zhao et al. 2002). The Mesoproterozoic was a Uralian margin; Bradley, 2008) while elsewhere, along the Grenville
time of few passive margins; at no time between 1750 and 1000 Ma orogen, Rodinia's assembly was still underway. The Rodinia supercon
was there a sudden increase in their abundance like that seen in the tinent cycle departed from the ideal.
Phanerozoic upon the breakup of Pangea. This suggests that Nuna
remained intact until at least 1000 Ma. 5.4. Gondwana
Relative quiescence during the tenure of a Mesoproterozoic
supercontinent provides an explanation for the “Boring Billion” of Whether or not the next continental assemblage Gondwana
Holland (2006). Secular records that show pronounced minima qualifies as a supercontinent depends on the definition, as discussed in
during the ca. 1600 1000 Ma interval include passive margins Section 2. Only if one adopts an inclusive definition, as I now do, does
(Fig. 11), detrital zircons (Figs. 15 and 16), seawater 87Sr/ 86Sr Gondwana qualify. Gondwana consisted of the cratons now in South
(Fig. 18), greenstone belt collisions (Fig. 19), eclogites (Fig. 20), America, Antarctica, Africa, Arabia, India, and Australia, as well as
granulites (Fig. 21), carbonatites (Fig. 22), and orogenic gold (Fig. 37). microcontinents in the Appalachian and Paleotethyan realms. Its fit is
well constrained because the now dispersed pieces can be brought
5.3. Rodinia together by closing the modern South Atlantic, Indian, and Southern
Oceans; the Gondwana fit is essentially the southern half of the Pangea fit.
The next supercontinent, Rodinia, is generally regarded as having In the geologically and paleomagnetically based animation of Li et
been assembled near the end of the Mesoproterozoic and disassembled al. (2008), collisional assembly of Gondwana is shown spanning the
in stages during the Neoproterozoic. Reconstructions of Rodinia are interval 690 to 530 Ma. This timing is consistent with the peak
notable for their lack of agreement. Differing interpretations of the abundance of passive margin end dates at 650 550 Ma (Fig. 13), with
position of West Africa were listed in Section 1; southwestern Laurentia the peak abundance of detrital zircons at 600 550 Ma (Fig. 16), and
is equally controversial, being variously positioned next to East with the peak abundance of orogenic granites at 588 Ma (Fig. 14).
Antarctica (Moores, 1991), Australia (Karlstrom et al., 2001), Siberia These secular trends conform to the patterns seen for Pangea. A key
(Sears and Price, 2003), South China (Li et al., 2008) or West Africa Plata point is that the assembly of Gondwana and disassembly of Rodinia
(Evans, 2009). Most reconstructions are variations on a theme in which overlapped in time. The final breakup of Rodinia, marked by the rift
two halves of Rodinia came together during the Grenville orogeny. In drift transition along the Ouachita passive margin, is dated at about
the very different model recently suggested by (Evans (2009), the 520 Ma (Bradley, 2008, and references therein).
Grenville orogen was not the result of a terminal, supercontinent Gondwana remained largely intact until colliding with Laurentia at
forming collision, but instead was an Andean type orogen that faced a about 300 Ma. Then it underwent a name change, becoming part of
Pacific type ocean, external to Rodinia. A common element in all these Pangea, which broke up starting ca. 180 Ma. Thus, the tenure of
reconstructions is the yet unproven assumption that most or all the Gondwana can be set at ca. 550 to ca. 180 Ma.
world's continents were part of a single supercontinent. Whether they
all were or not, the key point is that at least many of the continents were 6. Supercontinent cycles and other secular variation
together. This is evident from the initiation of new rifted margins on all
sides of Laurentia at about the time of the Neoproterozoic Cambrian Using the revised timetable (Fig. 46C) as a baseline, I next discuss
boundary (e.g., Hoffman, 1991). secular trends that bear on the evolving context of the succession of
Published assessments of Rodinia's dates are in general agreement. supercontinents and on the properties of particular cycles. The
Condie (2003a) suggested that it was assembled between 1300 and differences between cycles can be attributed to factors such as the
950 Ma, that it lasted from 950 to 850 Ma, and was disassembled from vagaries of plate geometry, long term decline of Earth's radiogenic
850 to 600 Ma. Similarly, Li et al. (2008) suggested that Rodinia was heat production, long term increase in viscosity of the residual mantle,
assembled between 1300 and 900 Ma, was at its largest from 900 to emergence of life as a major geochemical agent, and oxygenation of the
750 Ma, and was disassembled in stages from about 790 to 550 Ma (Li oceans and atmosphere. Many Earth processes are cyclic, for different
et al., 2008, animation). reasons and on different timetables, and interactions among different
I suggest a slightly different scenario and timetable based on the cycles have led to remarkable secular changes.
passive margin and zircon records. Abundances of detrital zircons
show a major peak at 1044 Ma and a subsequent low centered at 900 6.1. Global heat production and rates of plate tectonics
850 Ma (Fig. 16), consistent with the assembly of a Rodinia
supercontinent during Grenville age collisions. Comparable age The heat generated by the decay of 238U, 235U, 232Th, and 40K has
patterns are evident in the age distributions of greenstone belt exponentially declined through Earth history (Fig. 24, from Pollack,
collisions, eclogite and granulite facies metamorphism, and carbona 1997). The various curves correspond to different mixes of radioactive
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 29

fuel, with the “crustal” curve corresponding quite closely to Allegré et Atlantic type oceanic crust has generally lasted longer than it does on
al. (2001) assessment of the bulk composition of the Earth. Regardless Earth at present 3. Based on an equivocal stratigraphic record from
of the fuel mix, all published variations of this curve have the same Siberia, the oldest seafloor evidently endured for nearly 600 m.y.
basic form, showing an exponential decline in radiogenic heat before finally being subducted at about 1000 Ma (Bradley, 2008). In
production since the beginning. the sawtooth curve in Fig. 28, the longest lived margin at a given time
Mantle potential temperature, as inferred from the geochemistry ages along a line with slope of −1 until collision, when a different margin
of non arc basalts, shows a cooling trend over the latter half of Earth takes over as oldest. Steady increases from 2444 to 2130, 1600 to 1010,
history (Fig. 25; from Herzberg et al., 2010). Temperatures were as 849 to 558, and (less strikingly) 299 160 Ma correspond to tenures of
high as 1500 1600 °C at ca. 3000 2500 Ma, and have declined since. Sclavia and Superia (S S), Nuna (N), Rodinia (R), and Pangea (P).
Each successive supercontinent cycle thus played out on a cooler
planet. 6.3. Obducted ophiolites
Early Earth's extra heat is behind the widely held belief in the
geodynamics community that Precambrian global spreading rates Obducted ophiolites fragments of ancient oceanic lithosphere
were higher, and that the early Earth was occupied by smaller and caught up in orogenic belts have an uneven age distribution. Figs. 9
(or) faster plates (e.g., Burke et al., 1976; Hargraves, 1986; Pollack, and 29 (from Moores, 2002 and Dilek, 2003) show maxima at ca. 800
1997). Alternatively, based largely on paleomagnetic evidence, Kröner 750, 500 450, 160 150, and 100 90 Ma. The Ordovician peak largely
and Layer (1992) suggested that Archean rates of plate motion were reflects Appalachian forearc ophiolites (Cawood and Suhr, 1992) that
comparable to modern rates. More recently, Korenaga's (2006) were obducted during the earliest phases of Pangea's assembly. The
modeling of mantle evolution suggested an overall increase in global two Mesozoic peaks, which mostly correspond to ophiolites in the
spreading rate through the first 95% of Earth history; it has only Tethyan realm, are broadly synchronous with disassembly of Pangea.
declined since about 300 Ma (Fig. 26). The rationale is that higher The scarcity of ophiolites younger than 70 Ma rules out attrition as the
degrees of partial melting of the hotter mantle beneath ridges would main reason behind the age distribution. Whereas a single explana
have produced thicker oceanic crust (Fig. A30, from Moores, 2002) tion in terms of some aspect of the supercontinent cycle is elusive, it is
and a substantially more viscous residual mantle, resulting in slower clear that the vast majority of ophiolites formed and were obducted
global spreading rates than today (Korenaga, 2006). As supporting during some part of the Pangea cycle.
geologic evidence, Korenaga (2006) cited an apparent decrease in the
period between supercontinents over time. A new line of supporting 6.4. High pressure, low temperature metamorphic rocks
evidence and one that does not hinge on the subjective definition of
what is or isn't a supercontinent is provided by the longer lifespans High pressure, low temperature metamorphic rocks, classic in
of individual passive margins in the Proterozoic (Fig. 27; see dicators of plate convergence, form during subduction of both oceanic
Section 6.2) (Bradley, 2008). and continental crust. Fig. 30 shows the distribution of blueschists and
Shorter term fluctuations in global spreading rates are another ultra high pressure metamorphic rocks through time (Brown, 2007).
matter. The Late Cretaceous, in particular, has been widely regarded as The oldest blueschists date from assembly of Gondwana (ca. 680 Ma
a time of faster spreading rates (Hays and Pitman, 1973), although a blueschists from Brazil; Parkinson et al., 2001). In the Phanerozoic
recent reappraisal by Rowley (2002) found no supporting evidence in (Fig. A2), blueschists and ultra high pressure metamorphic rocks are
the slim magnetic anomaly record. Slower global spreading rates are relatively abundant from ca. 400 to 300 during assembly of Pangea,
to be expected during times of supercontinent tenure (Silver and and from ca 100 50 Ma during disassembly of Pangea. They are
Behn, 2008), but there is no seafloor record with which to directly test relatively scarce from 300 to 150 Ma, during Pangea's tenure (Fig. A2).
this idea. Thus, their age distribution is broadly consistent with the Pangea
Additional secular trends bearing on the physics and chemistry of supercontinent cycle.
the mantle are shown in Figures A30, A31, A33, A36, A38, A40, and The greatest metamorphic pressures are seen in Phanerozoic rocks
A41. (Fig. 31, from Brown, 2008) whereas the steepest inferred geothermal
gradients, based on sparse data, are seen in Paleoproterozoic rocks
6.2. Oceanic crust and passive margin proxies (Fig. 32, from Brown, 2008). The most straightforward interpretation
is that blueschists and related rocks are absent from the record before
The age distribution of oceanic crust is akin to that of a human the Neoproterozoic because P T conditions in subduction zones were
population, with spreading like birth and subduction like death different.
(Veizer and Jansen, 1979, 1985; Veizer et al., 1989; Veizer and
Mackenzie, 2003). The proximal cause of this age pattern (Fig. 8; from 6.5. Mantle derived igneous rocks
Sclater et al., 1980) is attrition: only survivors are counted. Based only
on the current situation, oceanic crust would appear to have an Secular cooling of the mantle (Section 6.1) is broadly consistent
oblivion age of about 180 Ma, a half life of about 60 m.y., and a with the age distribution of the ultramafic volcanic rock, komatiite. A
recycling rate of about 3.5 km 2/yr (Veizer and Mackenzie, 2003). probability density plot (Fig. 33; constructed from Table 1 of Isley and
These particular values, however, are contingent on our current place Abbott, 1999) shows peak abundances of komatiites at ca. 3400 and
in the supercontinent cycle. Thirty million years from now, the 2750 Ma, and a single occurrence in the Cretaceous. Figure A40 shows
Atlantic will have widened, the apparent half life of the world's the abundance of komatiite relative to other volcanic rocks in the
oceanic crust will be longer, and the apparent recycling rate slower. greenstone belts in which they occur (from Condie and O'Neill, 2010),
For the deeper past we can glean anecdotal information from the after a similar plot by de Wit and Ashwal, 1997); the ratio confirms
passive margin compilation described in Section 4.2. The lifespan of a that the prominent peak at ca. 2750 Ma in Fig. 33 is not merely an
passive margin (Fig. 27), namely its start date minus its end date artifact of preservation. Potential temperatures for komatiites are
(Bradley, 2008), is dictated by the rates of plate motion during substantially higher than for modern basalts, roughly 1600 1800 °C
opening and closing phases of the ocean basin. These dates can be (Herzberg et al., 2010) (Fig. 25). Komatiites have been most widely
recast to track the age of the oldest Atlantic type seafloor through interpreted as the products of mantle plumes (Ernst and Buchan,
time (Fig. 28; see Supplementary data for explanation); this works
because of the equivalence between the lifespan of a passive margin 3
This analysis does not bear on the longevity of Pacific-type oceanic crust through
and the age (at time of collision) of the oldest seafloor that bordered it. time, and thus tells only part of the story.
 Author's personal copy

30 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

2001; Isley and Abbott, 2002), although a subduction origin has also of Gondwana or Pangea. The overall age distribution is therefore
been advocated (Grove and Parman, 2004). Regardless, the early somewhat problematic.
abundance of komatiites implies that the mantle was hotter than
today and that by ca. 1700 it had cooled sufficiently that komatiites 6.8. Sedimentary rocks
became exceedingly rare. This explains their occurrence in orogens
related to assembly of Vaalbara, Superia, Sclavia, and Nuna and their Several compilations have shown that the global abundance of
absence in orogens formed during assembly of Rodinia, Gondwana, or sedimentary rock decreases back through time. In a detailed plot of
Pangea. just the Phanerozoic (Fig. 10) (Veizer and Mackenzie, 2003, who
The neodymium isotopic system, based on the decay of 147Sm to attributed the plot to W.W. Hay), sedimentary rocks are subdivided
143
Nd, is a good tracer of both crustal and mantle evolution because into three depositional settings: epicontinental, passive margin, and
Sm and Nd fractionate during partial melting at mantle depths. 4 The ocean basin. The abundance of geologically young passive margin and
neodymium isotopic composition of the depleted mantle is thus more ocean basin strata can be attributed to three effects: (1) attrition the
radiogenic (positive εNd) than that of the continental crust (negative geologic destruction of these tectonic settings over time; (2) rules of
εNd). Fig. 34 (from Bennett, 2003) tracks εNd of proxies for the the census, which discounted older oceanic or passive margin strata
depleted mantle over time, specifically modern MORBs, Phanerozoic after they have been involved in orogeny; and (3) our current place
ophiolites, and Precambrian juvenile granitoids. The data show a clear about 180 m.y. into a supercontinent cycle.
secular increase in εNd of the depleted mantle, toward ever more Unfortunately, there is no comparable plot of sedimentary rock
radiogenic values. The general clustering of analyses from around volume through the span of Earth history. The best available appears
2700 2500, 2100 1800, and 600 500 Ma is probably a reflection of to be compilation of Ronov et al. (1991) (Fig. 38), which stopped well
the relative abundance of rocks of these ages. Each supercontinent short, leaving out the Paleoproterozoic and Archean. This shows an
formed in the context of ever more depleted mantle. The data are yet overall decline back through time: younger strata are more abundant
too sparse to show whether the overall increase was steady, or because older strata have been eroded away, subducted, or so
punctuated by the supercontinent cycle. tectonized that they are no longer counted as sedimentary rocks.
The subcontinental lithospheric mantle evolved as the succession Fluctuations that might correlate with older supercontinent cycles
of supercontinents came and went. Mantle xenoliths and xenocrysts cannot be detected in this time averaged dataset.
from beneath crustal provinces of various ages show systematic
variations through time in density (Fig. 35), mineralogy (Fig. A30), 6.9. Sedimentary recycling
and chemistry (Fig. A31) (Poudjom Djomani et al., 2001; Griffin et al.,
2003; Griffin et al., 2009). Archean cratons have deeper, more Veizer and Mackenzie (2003, and older references therein)
buoyant, more depleted keels than Proterozoic or Phanerozoic estimated that the Earth's sedimentary mass is ~90% recycled from
continents. The age resolution of these findings is still too coarse to older sedimentary rock. The time scale of recycling is revealed by the
reveal any secular variations on the time scale of individual neodymium isotopic composition of shales and sandstones (Fig. 39,
supercontinent cycles. from Vervoort et al., 1999), and by detrital zircons (Fig. 40). The x axis
in both plots is the depositional age. In Fig. 39, the y axis is the crustal
6.6. Massif type anorthosites residence age, i.e. the time since the continental crust from which the
sediments were derived originally separated from the mantle. In
Massif type anorthosites plutons having N90% plagioclase have Fig. 40, a new plot, the y axis is the igneous age of detrital zircons,
an unusual age distribution (Fig. 36) featuring a broad high from about typically 60 to 100 ages per sandstone sample. Both plots show that,
2000 to 1000 Ma, roughly the time of Holland's (2006) “Boring Billion”. relative to the time of deposition, sedimentary grains are older now
Not unexpectedly, a similar age distribution is shown by anorthosite than in the past. Thus, a sandstone deposited during the assembly of
hosted titanium ores (Meyer, 1988; not shown). Massif type anortho Nuna consisted of then newer clastic material than a sediment
sites first appeared in abundance during the assembly of Nuna and deposited during the assembly of Pangea. This makes it harder and
virtually disappeared after the assembly of Rodinia. The population is harder to create so called “juvenile crust” (cf. Fig. A27) during each
dominated by anorthosites from Laurentia and Baltica; the tectonic successive supercontinent cycle.
setting of emplacement appears to have been along a long lived
convergent margin (Ashwal, 1993). Hoffman (1989) suggested that the 6.10. Oxygenation of the atmosphere and oceans
Laurentian anorthosites formed at a time of greater global heat
production when Nuna sat almost immobile with respect to the mantle, Secular trends identify three tipping points in the oxygenation
a process that no longer can occur because of secular cooling. history of the exosphere, at ca. 2500, 1850, and 580 Ma. The existence
of a “Great Oxidation Event” (or GOE) near the Archean Proterozoic
6.7. Orogenic gold deposits boundary has long been suspected and is now well documented by
means of sulfur isotopes (e.g., Farquhar et al., 2010). Fig. 41 shows
Orogenic gold deposits are products of plate convergence, being Δ 33S in a variety of sulfide and sulfate minerals is plotted as a function
formed at mid crustal levels during regional prograde metamorphism of time, from a compilation by Domagal Goldman et al. (2008). Δ 33S is
(Goldfarb et al., 2001). Peak abundances are seen at 2700 2600 and a measure of mass independent fractionation sulfur isotopes. An
200 100 Ma (Fig. 37). Few orogenic gold deposits would be expected abrupt, dramatic transition occurred from a range of both negative
at times of tectonic quiescence, during the tenures of supercontinents. and positive values before ca. 2500 Ma, to near zero since then. The
Consistent with this prediction, since the Neoarchean, there are two isotopes of sulfur fractionate independently of mass when solar UV
major intervals when orogenic gold deposits are unknown: ca. 2500 radiation strikes sulfur bearing dust in the atmosphere. Ozone in the
2200 and ca. 1700 800 Ma. The older interval corresponds to the stratosphere blocks solar UV radiation so its presence (and by
tenures of Sclavia and Superia (or of Kenorland); the younger interval extension, the presence of O2 in the troposphere below) can be
corresponds to the combined tenures of Nuna and Rodinia. Orogenic inferred starting at ca. 2500 Ma.
gold deposits, however, are not particularly sparse during the tenures The age distribution of Superior type banded iron formations
(BIFs) reveals another transition at ca. 1850 Ma. BIFs were widely
4
The hafnium isotopic system, based on the decay of 176Lu to 176Hf, is comparable to deposited between about 3000 Ma and about 1850 Ma, vanished from
the neodymium system (Fig. A33). the record for about more than a billion years, made a brief resurgence
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 31

associated with Neoproterozoic glaciations, and vanished again Appalachian, and Ouachita orogens. Similarly, the preceding major
(Bekker et al., 2010) (Fig. 42). Superior type BIFs accumulated in lowstand, at ca. 550 Ma (Fig. 7), correlates with the start of
platformal to outer ramp settings and many, at least, are associated Gondwana's tenure (Condie, 2005).
with the terminal collisions of long lived passive margins (Hoffman, The supercontinent cycle is not the only driver of sea level
1987). Precipitation of magnetite (the principal iron mineral in BIFs) fluctuations. Others include glaciation (a contributor to the present
from seawater requires reducing conditions and low concentrations of day low) and the distribution of continents and ocean basins over
sulfur (Huston and Logan, 2004). Evidently, such conditions did not geoid highs and lows. Proterozoic sea level fluctuations are still too
pertain before ca. 3200 Ma or after about 1850 Ma. The apparently poorly resolved to bear definitively on older supercontinent cycles.
simultaneous global demise of BIFs at 1850 Ma coincides with the
Sudbury impact event in Canada (Slack and Cannon, 2008). 6.13. Mississippi Valley type lead zinc deposits
The concentration of molybdenum in marine black shales
suggests a third transition at ca. 580 Ma. Before that time, Mo Mississippi Valley type (MVT) lead zinc deposits are epigenetic
concentrations ranged from a few to about 70 ppm; after 580 Ma, lead zinc ores hosted in carbonate rocks (e.g. Leach et al., 2010). Their
Mo concentrations jumped by an order of magnitude (Fig. 43; Scott distribution through Earth history, as measured by amount of
et al., 2008). Scott et al. (2008) inferred that this transition dates the contained metals (Fig. 45) is dominated by Phanerozoic deposits
oxidation of the deep ocean and advent of the redox and nutrient and particularly by ones associated with the tenure of Pangea. Most
cycles that pertain today. MVT deposits are in passive margin platform sequences now located
Campbell and Allen (2008) have suggested that these transitions in fold thrust belts or orogenic forelands (Bradley and Leach, 2003). In
in the behaviors of sulfur, iron, and molybdenum were the result of the type area in the south central United States, MVT host carbonates
supercontinent assembly. In this scenario, orogeny during supercon were laid down along a Cambrian to Mississippian passive margin that
tinent assembly supplies nutrients such as phosphorus and iron to the originated during the last fragmentation of Rodinia. The margin was
oceans in greater than normal amounts. This increases photosynthesis inundated by synorogenic clastics during the Ouachita orogeny, one of
and oxygen production while also promoting burial of organic carbon. the last of the Pangea forming collisions, in the Pennsylvanian. The
deposits themselves were formed in the orogenic foreland from a
6.11. Glaciation continent scale hydrothermal system that recharged in the orogen.
Thus the host rocks are from the dispersal phase of one superconti
Glacial intervals in Earth history took place during the assembly, nent cycle and the mineralizing event was part of the assembly phase
tenure, and collision stages of various supercontinent cycles (Fig. 44). of a second supercontinent cycle. As is the case for many types of ore
The Pleistocene ice sheets spread across the widely dispersed deposits. MVT formation requires the confluence of other contributing
fragments of Pangea, long after its breakup. In constrast, the Paleozoic factors as well. Specifically: (1) MVT deposits form from brines in the
ice sheets of the southern hemisphere existed during times of arid latitudinal belts at ca. 15 30°N and S (Leach et al., 2010 this is a
supercontinent tenure: in the Pennsylvanian to Permian on Pangea, nearly untapped exploration criterion); and (2) mineralizing brines
and in the Ordovician and Devonian on Gondwana. The regional must contain free oxygen, ruling out the existence of this type of
Gaskiers glaciation at ca. 580 Ma took place during the overlapping deposit in the Archean.
time of Rodinia's breakup and Gondwana's assembly. This is also the
case for the global Marinoan and Sturtian glaciations at ca. 635 and ca. 6.14. Other secular trends
700 Ma. The Paleoproterozoic (ca. 2300 2200 Ma) Gowganda global
glaciation correlates in general with the tenures of Sclavia and The plots in Figs. 3 45 bear on the supercontinent timetable, the
Superia. In a review of tectonic settings of glaciogenic rocks, Eyles evolving context of the succession of supercontinents, the properties
(2008) found a preponderance that he interpreted as rift related. of particular cycles, and long term background trends. Figures A1 A86
in the Supplementary data section show additional secular trends that
6.12. Sea level were compiled for this study. They pertain to the continental crust,
the mantle, granitoid chemistry, seawater chemistry, sedimentary
Some first order sea level changes can be correlated with the chemistry, sedimentary rock types, the fossil record, ore deposits, and
Pangea supercontinent cycle. Sea level as recorded by the proportion extraterrestrial variables. The number of recently published secular
of platform flooding was generally high in the early and middle trends of global scope attests to the growth of this approach to Earth
Paleozoic, low in the late Paleozoic to early Mesozoic, high during the history.
Cretaceous, and relatively low today (Fig. 7, from Hallam, 1992). The
broad low closely matches the tenure of Pangea as indicated by the 7. Summary
area of the largest continent (Fig. 3) (Worsley et al., 1984, 1986). This
correlation can be explained by a model proposed by Heller and Two complementary approaches to Earth history plate recon
Angevine (1985) that relates sea level change to ridge volume and to structions and analysis of secular trends have been brought together
the supercontinent cycle. Continental breakup is followed by growth in this assessment of the supercontinent cycle. A global plate
of an Atlantic type ocean at the expense of a Pacific type ocean, i.e., reconstruction for a particular time provides a unified rationale for
one that is bounded by subduction zones. At first, the new Atlantic the origins of, interrelations between, and geographic distribution of
type seafloor is younger than the average seafloor that is necessarily the rocks of that age. In general, this approach gets harder and less
subducted elsewhere, so sea level rises. Later, the growing Atlantic certain back through time. A few intervals in the Precambrian,
type ocean becomes older than the average subducting seafloor, so sea however, are more tractable: the supercontinent intervals. But each
level falls. Thus a lag is built into the process, such that the highstand Precambrian supercontinent is subject to debate as to timing of
follows initial opening by 70 to 120 m.y. (Heller and Angevine, 1985). assembly and breakup, configuration of the pieces, relative size, and
Some of the finer texture of the Phanerozoic sea level curve may be even whether or not the term “supercontinent” should be applied.
due to the Heller and Angevine (1985) mechanism applied to The supercontinent cycle appears to be manifested in a number of
individual ocean basins and individual Wilson cycles, each operating secular trends. It is well suited to study by analysis of multiple trends
on slightly different timetables. Low sea level at the start of Pangea's because plate reorganizations, which necessarily take place when
tenure can be linked to the requisite death or slowing of mid ocean ocean basins start to open and when they finish closing (Silver and
ridges when convergence ceased across the Central Asian, Uralian, Behn, 2008), influence many parts of the Earth system. For present
 Author's personal copy

32 D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33

purposes, the most informative trends are those that in some way used to construct Fig. 14 and A43 A48, and Christ Hawkesworth and
track the assembly, tenure, and disassembly of Pangea. The tenure of Bruno Dhuime provided a high resolution version of Figure A36.
Pangea was a time of few passive margins and a relative low in zircon
production. Followed deeper into the past, these two variables Appendix A. Supplementary data
together suggest that the tenures of putative Precambrian supercon
tinents coincided with zircon minima at ca. 2450 2225 (Superia and Supplementary data to this article can be found online at doi:10.
Sclavia, or Kenorland), ca. 1625 1000 (Nuna), and ca. 875 725 Ma 1016/j.earscirev.2011.05.003.
(Rodinia). Other secular trends that show this same pulse include the
abundances of granulites, eclogites, carbonatites, volcanogenic mas
References
sive sulfides, and greenstone belt collisions.
The supercontinent cycle has been characterized as periodic (e.g. Allegré, C., Manhes, G., Lewin, E., 2001. Chemical composition of the Earth and the
Korenaga, 2006). Indeed this is the obvious inference from older volatility control on planetary genetics. Earth and Planetary Science Letters 185,
49–69.
compilations such as Condie's (1997) plot of orogenic granites (Fig.
Ashwal, L.D., 1993. Anorthosites. Springer-Verlag, Berlin. 422 p.
A21), which shows three remarkable peaks in abundance at 2700 2600, Barley, M.E., Groves, D.I., 1992. Supercontinent cycles and the distribution of metal
1900 1800, and 1100 1000 Ma. The case for true periodicity is not so deposits through time. Geology 20, 291–294.
compelling if one judges from more recent compilations of zircon Bekker, A., Slack, J.F., Planavsky, N., Krapež, B., Hofmann, A., Konhauser, K.O., Rouxel,
O.J., 2010. Iron formation: the sedimentary product of a complex interplay among
abundances (Figs. 14, 15, and 16), which show a number of minor mantle, tectonic, oceanic, and biospheric process. Economic Geology 105,
maxima and minima in addition to the main ones. How these lesser 467–508.
fluctuations relate to supercontinent assembly, tenure, and disassembly Bennett, V.C., 2003. Compositional evolution of the mantle. Treatise on Geochemistry 2,
493–519.
ties into whether or not a relatively small grouping of continents, like Bleeker, W., 2003. The late Archean record: a puzzle in ca. 35 pieces. Lithos 71, 99–134.
Gondwana, counts as a supercontinent. This is not merely a question of Bradley, D.C., 2008. Passive margins through earth history. Earth-Science Reviews 91,
semantics; it underscores the need for a quantitative measure of 1–26.
Bradley, D.C., Leach, D.L., 2003. Tectonic controls of Mississippi Valley-type lead-zinc
“supercontinentality.” mineralization in orogenic forelands. Mineralium Deposita 38, 652–667.
Each supercontinent cycle was unique for multiple reasons. On Bradley, D.C., Dewey, J.F., 2005. Penrose conference report: secular variation in
Earth today we can see how intricate the shape and layout of the tectonics and allied fields. GSA Today 15 (3), 12.
Brown, M., 2007. Metamorphic conditions in orogenic belts: a record of secular change.
continents can be, and how complicated the mosaic of boundaries
International Geology Review 49, 193–234.
between plates, large and small. There is no reason to expect that Brown, M., 2008. Characteristic thermal regimes of plate tectonics and their
things were ever much simpler during earlier times when continents metamorphic imprint throughout Earth history: when did Earth first adopt a
plate tectonics mode of behavior? Geological Society of America Special Paper 440,
were dispersed. Each supercontinent cycle was also unique because of
97–128.
its time in Earth history. The successive supercontinent cycles played Burke, K., Dewey, J.F., Kidd, W.S.F., 1976. Dominance of horizontal movements, arc
out against inexorable long term changes related to Earth's declining and microcontinental collisions during the later Permobile Regime. In: Windley,
radiogenic heat production. As the mantle cooled, the degree of partial B.F. (Ed.), The Early History of the Earth. John Wiley & Sons, New York, pp.
113–130.
melting beneath ridges decreased, and the viscosity of the residual Burke, K., Ashwal, L.D., Webb, S., 2003. New way to map old sutures using deformed
mantle accordingly decreased. Taking this into account, Korenaga's alkaline rocks and carbonatites. Geology 31, 391–394.
(2006) modeling results suggest that through the Proterozoic, plate Campbell, I.H., Allen, C.M., 2008. Formation of supercontinents linked to increases in
atmospheric oxygen. Nature Geoscience 1, 554–558. doi:10.1038/ngeo259.
motions sped up, and that through the Phanerozoic, they slowed Cawood, P.A., Suhr, G., 1992. Generation and obduction of ophiolites: constraints from
down. Once common rock types went virtually extinct: banded iron the Bay of Islands Complex, western Newfoundland. Tectonics 11, 884–897.
formation, komatiite, and massif type anorthosite. Other rock types Cogley, J.G., 1984. Continental margins and the extent and number of continents.
Reviews of Geophysics and Space Physics 22, 101–122.
became common, or came into being: blueschist, coal, and bioclastic Collins, W.J., 2003. Slab pull, mantle convection, and Pangaean assembly and dispersal.
limestone. Through it all, sea level fluctuated via a complex interplay Earth and Planetary Science Letters 205, 225–237.
of ridge volume, glaciation, and positions of ocean basins and Condie, K.C., 1994. Greenstones through time. Developments in Precambrian Geology
11, 85–120.
continents over geoid highs and lows. Only the Pangea cycle can be
Condie, K.C., 1997. Plate Tectonics and Crustal Evolution, 4th edition. Butterworth-
understood in the context of Earth processes as they operate today. Heineman, Boston. 282 p.
Every secular trend can have only one correct explanation, Condie, K.C., 2003. Supercontinents, superplumes and continental growth: the
Neoproterozoic record. Geological Society of London Special Publications 206, 1–21.
complicated though it might be. Moreover, because the Earth has
Condie, K.C., 2005. Earth as an Evolving Planetary System. Elsevier Academic Press,
only one history, any viable explanation for one secular trend must Amsterdam. 447 p.
honor all the rest. An encyclopedic approach is indicated. Condie, K.C., O’Neill, C., 2010. The Archean-Proterozoic boundary: 500 My of tectonic
transition in Earth history. American Journal of Science 310, 775–790.
Condie, K.C., Belousova, E., Griffin, W.L., Sircombe, K.N., 2009a. Granitoid events in space
and time: constraints from igneous and detrital zircon age spectra. Gondwana
Acknowledgments Research 15, 228–242.
Condie, K.C., O'Neill, C., Aster, R.C., 2009b. Evidence and implications for a widespread
magmatic shutdown for 250 My on Earth. Earth and Planetary Science Letters 282,
I am grateful to reviewers Alison Till, Francis Macdonald, Kent 294–298.
Condie, Rich Goldfarb, and Alfred Kröner for their many constructive Condie, K.C., Bickford, M.E., Aster, R.C., Belousova, E., Scholl, D.W., 2011. Episodic zircon
suggestions. This paper grew out of the 2004 Penrose Conference on ages. Hf isotopic composition and the preservation rate of continental crust.
Geological Society Bulletin. 123, 951–957.
Secular Variation in Tectonics and Allied Fields (Bradley and Dewey, de Wit, M.J., and Ashwal, L.D., 1997. Convergence towards divergent models of
2005), which received support from the USGS Venture Capital Fund. greenstone belts. In: M.J. de Wit and L.D. Ashwal (Eds.), Greenstone Belts: Oxford
The global perspective needed for this kind of synthesis came from University Press, New York, p. ix–xix.
Dewey, J.F., 2007. The secular evolution of plate tectonics and the continental crust: an
Kevin Burke and John Dewey. Theresa Taylor, Heather Bleick, and Sam outline. Geological Society of America Memoir 200, 1–7. doi:10.1130/2007.1200(01).
Friedman helped with the detrital zircon compilation. I benefited from Dewey, J.F., Burke, K., 1974. Hotspots and continental breakup: implications for
various discussions and exchanges with Wouter Bleeker, Mike Brown, collisional orogeny. Geology 2, 57–60.
Dilek, Y., 2003. Ophiolite pulses, orogeny, and mantle plumes. Geological Society of
Bill Collins, Puol Emsbo, David Evans, Rich Goldfarb, Peter Haeussler,
London Special Publications 218, 9–19.
Paul Hoffman, Karl Karlstrom, Dave Leach, Andrew McCauley, Joe Domagal-Goldman, S.D., Kasting, J.F., Johnston, D.T., Farquhar, J., 2008. Organic haze,
Meert, Greg Retallack, Celal Sengör, Graham Shields, Paul Silver, Bob glaciations and multiple sulfur isotopes in the Mid-Archean Era. Earth and
Stern, Jan Veizer, Maarten de Wit, and many others. Gerard Stampfli Planetary Science Letters 269, 29–40.
Evans, D.A.D., 2009. The paleomagnetically viable, long-lived and all-inclusive Rodinia
provided the plate reconstructions in Fig. 1; Ian Campbell provided supercontinent reconstruction. Geological Society, London, Special Publications
data tables used to construct Fig. 15; Kent Condie provided data tables 327, 371–404.
 Author's personal copy

D.C. Bradley / Earth-Science Reviews 108 (2011) 16–33 33

Eyles, N., 2008. Glacio-epochs and the supercontinent cycle after ∼3.0 Ga: tectonic Murphy, J.B., Nance, R.D., 2007. Do supercontinents turn inside-in or inside-out?
boundary conditions for glaciation. Palaeogeography, Palaeoclimatology. Palaeoe- International Geology Review 47, 591–619.
cology 258, 89–129. Nance, R.D., Worsley, T.R., Moody, J.B., 1986. Post-Archean biogeochemical cycles and
Farquhar, J., Wu, N., Canfield, D.E., Oduro, H., 2010. Connections between sulfur cycle long-term episodicity in tectonic processes. Geology 14, 514–518.
evolution, sulfur isotopes, sediments, and base metal sulfide deposits. Economic Parkinson, C.D., Motoki, A., Onishi, C.E., Maruyama, S., 2001. Ultrahigh-pressure pyrope-
Geology 105, 509–533. kyanite granulites and associated eclogites in Neoproterozoic nappes of Southeast
Garrels, R.M., Mackenzie, F.T., 1971. Evolution of Sedimentary Rocks. Norton, New York. Brazil. [abs.] UHPM Workshop 2001, Fluid/slab/mantle interactions and ultrahigh-P
397 p. minerals. Waseda University, Tokyo, pp. 87–90. Abstract Volume.
Goldfarb, R.J., Groves, D.I., Gardoll, S., 2001a. Orogenic gold and geologic time: a global Pollack, H.N., 1997. Thermal characteristics of the Archean. In: De Wit, M.J., Ashwal, L.D.
synthesis. Ore Geology Reviews 18, 1–75. (Eds.), Greenstone Belts. Clarendon Press, Oxford, pp. 223–232.
Goldfarb, R.J., Bradley, D.C., Leach, D.L., 2010. Secular variation in economic geology. Poudjom Djomani, Y.H., O'Reilly, S.Y., Griffin, W.L., Morgan, P., 2001. The density
Economic Geology 105, 459–465. structure of subcontinental lithosphere through time. Earth and Planetary Science
Gradstein, F.M., Ogg, J.G., 2004. A geologic time scale 2004. Cambridge University Press, Letters 184, 605–621.
Cambridge. 598 p., 1 plate. Prokoph, A., Ernst, R.E., Buchan, K.L., 2004. Time-series analysis of Large Igneous
Griffin, W.L., O'Reilly, S.Y., Abe, N.S., Aulbach, S., Davies, R.M., Pearson, N.J., Doyle, B.J., Provinces: 3500 Ma to present. Journal of Geology 112, 1–22.
Kivi, K., 2003. The origin and evolution of Archean lithospheric mantle. Reddy, S.A., Evans, D.A.D., 2009. Palaeoproterozoic supercontinents and global
Precambrian Research 127, 19–41. evolution: correlations from core to atmosphere. Geological Society, London,
Griffin, W.L., O'Reilly, S.Y., Afonoso, J.C., Begg, G.C., 2009. The composition and evolution Special Publication 323, 1–26.
of lithospheric mantle: a re-evaluation and its tectonic implications. Journal of Richter, F.M., Rowley, D.B., DePaolo, D.J., 1992. Sr isotopic evolution of seawater: the
Petrology 50, 1185–1204. role of tectonics. Earth and Planetary Science Letters 109, 11–23.
Grove, T.L., Parman, S.W., 2004. Thermal evolution of the Earth as recorded by Rino, S., Kon, Y., Sato, W., Maruyama, S., Santosh, M., Zhao, D., 2008. The Grenvillian and
komatiites. Earth and Planetary Science Letters 219, 173–187. Pan-African orogens: world's largest orogenies through geologic time, and their
Groves, D.I., Condie, K.C., Goldfarb, R.J., Hronsky, J.M.A., Vielreicher, R.M., 2005. Secular implications on the origin of superplume. Gondwana Research 14, 51–72.
changes in global tectonic processes and their influence on the temporal Rogers, J.J.W., Santosh, M., 2002. Configuration of Columbia, a Mesoproterozoic
distribution of gold-bearing mineral deposits. Economic Geology 100, 203–224. supercontinent. Gondwana Research 5, 5–22.
Hallam, A., 1992. Phanerozoic Sea-Level Changes. Columbia University Press, New York. Rogers, J.J.W., Santosh, M., 2003. Supercontinents in Earth history. Gondwana Research
266 p. 6, 357–368.
Haq, B.U., Schutter, S.R., 2008. A chronology of Paleozoic sea-level changes. Science 322, Ronov, A.B., Yaroshevskiy, A.A., Migdisov, A.A., 1991. Chemical constitution of the
64–68. doi:10.1126/science.1161648. Earth's crust and geochemical balance of the major elements (Part I). International
Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Geology Review 33, 941–1097.
Triassic. Science 235 (4793), 1156–1167. doi:10.1126/science.235.4793.1156. Rowley, D.B., 2002. Rate of plate creation and destruction: 180 Ma to present.
Hargraves, R.B., 1986. Faster spreading or greater ridge length during the Archean? Geological Society of America Bulletin 114, 927–933.
Geology 14, 750–752. Rukhlov, A.S., Bell, K., 2010. Geochronology of carbonatites from the Canadian and
Hawkesworth, C.J., Dhuime, B., Pietranik, A.B., Cawood, P.A., Kemp, A.I.S., Storey, C.D., Baltic Shields, and the Canadian Cordillera: clues to mantle evolution. Mineralogy
2010. The generation and evolution of the continental crust. Journal of the and Petrology 98, 11–54. doi:10.1007/s00710-009-0054-5.
Geological Society, London 167, 229–248. doi:10.1144/0016-76492009-072. Sclater, J.G., Jaupart, C., Galson, D., 1980. The heat flow through oceanic and continental crust
Hays, J.D., Pitman III, W.C., 1973. Lithospheric plate motion, sealevel changes and and the heat loss of the Earth. Reviews of Geophysics and Space Physics 18, 269–311.
climatic and ecological consequences. Nature 246, 18–22. Scotese, C.R., 1997. Continental Drift Flip Book, 7th edition. Arlington, Texas. 80 p.
Heller, P.L., Angevine, C.L., 1985. Sea-level cycles and the growth of Atlantic-type Scott, C., Lyons, T.W., Bekker, A., Shen, Y., Poulton, S.W., Chu, X., Anbar, A.D., 2008.
oceans. Earth and Planetary Science Letters 75, 417–426. Tracing the stepwise oxygenation of the Proterozoic ocean. Nature 452, 456–459.
Herzberg, C., Condie, K., Korenaga, J., 2010. Thermal history of the Earth and its Sears, J.W., Price, R.A., 2003. Tightening the Siberian connection to western Laurentia.
petrological expression. Earth and Planetary Science Letters 292, 79–88. Geological Society of America Bulletin 115, 943–953.
Hoffman, P.F., 1987. Early Proterozoic foredeeps, foredeep magmatism, and superior- Sengör, A.M.C., Altiner, D., Cin, A., Ustaömer, T., Hsu, K.J., 1988. Origin and assembly of
type iron-formation of the Canadian shield. American Geophysical Union the Tethyside orogenic collage at the expense of Gondwana Land. Geological
Geodynamics Series 17, 85–98. Society of London Special Publication 37, 119–181.
Hoffman, P.F., 1989. Speculations on Laurentia's first gigayear (2.0 to 1.0 Ga). Geology Shields, G.A., 2007. A normalised seawater strontium isotope curve: possible
17, 135–138. implications for Neoproterozoic–Cambrian weathering rates and the further
Hoffman, P.F., 1991. Did the breakout of Laurentia turn Gondwana inside out? Science oxygenation of the Earth. eEarth 2, 35–42.
252, 1409–1412. Shields, G., Veizer, J., 2002. Precambrian marine carbonate isotope database: version
Hoffman, P.F., 1997. Tectonic genealogy of North America. In: van der Pluijm, B.A., 1.1. Geochemistry, Geophysics, Geosystems 6, 1–12.
Marshak, S. (Eds.), Earth Structure: an Introduction to Structural Geology and Silver, P.G., Behn, M.D., 2008. Intermittent plate tectonics. Science 319, 85–88.
Tectonics. McGraw-Hill, New York, pp. 459–464. Slack, J.F., Cannon, W.F., 2008. Shifts in global metallogeny at 1850 Ma: a connection to
Hoffman, P.F., 1999. The break-up of Rodinia, birth of Gondwana, true polar wander and the giant Sudbury impact event? Geological Society of America Abstracts with
the snowball Earth. Journal of African Earth Sciences 17, 17–33. Programs 40 (6), 146.
Hoffman, P.F., 2009. Pan-Glacial-a third state in the climate system. Geology Today 25, Stampfli, G.M., Borel, G.D., 2002. A plate tectonic model for the Paleozoic and Mesozoic
100–107. constrained by dynamic plate boundaries and restored synthetic oceanic isochrons.
Huston, D.L., Logan, G.A., 2004. Barite, BIFs and bugs: evidence for the evolution of the Earth and Planetary Science Letters 196, 17–33.
Earth's early hydrosphere. Earth and Planetary Science Letters 220, 41–55. Stampfli, G., Marcoux, J., Baud, A., 1991. Tethyan margins in space and time.
Isley, A.E., Abbott, D.H., 1999. Plume-related mafic volcanism and the deposition of Palaeogeography, Palaeoclimatology, Palaeoecology 87, 373–409.
banded iron formation. Journal of Geophysical Research 104, 5,461–15,477. Veizer, J., Jansen, S.L., 1979. Basement and sedimentary recycling and continental
Isley, A.E., Abbott, D.H., 2002. The intensity, occurrence, and duration of superplume evolution. Journal of Geology 87, 341–370.
events and eras over geologic time. Journal of Geodynamics 34, 265–307. Veizer, J., Jansen, S.L., 1985. Basement and sedimentary recycling: 2. Time dimension to
Karlstrom, K.E., Ahall, K.-I., Harlan, S.S., Williams, M.L., McClelland, J., Geissman, J.W., global tectonics. Journal of Geology 93, 625–643.
2001. Long-lived (1.8-1.0 Ga) convergent orogen in southern Laurentia, its Veizer, J., Mackenzie, F.T., 2003. Evolution of sedimentary rocks. Treatise on
extensions to Australia and Baltica, and implications for refining Rodinia. Geochemistry 7, 369–407.
Precambrian Research 111, 5–30. Veizer, J., Laznicka, P., Jansen, S.L., 1989. Mineralization through geologic time:
Korenaga, J., 2006. Archean geodynamics and thermal evolution of Earth: Archean recycling perspective. American Journal of Science 289, 484–524.
geodynamics and environments. AGU Geophysical Monograph Series 164, 7–32. Vervoort, J.D., Patchett, P.J., Blichert-Toft, J., Albarede, F., 1999. Relationships between
Kröner, A., Layer, P., 1992. Crust formation and plate motion in the early Archean. Lu–Hf and Sm–Nd isotopic systems in the global sedimentary system. Earth and
Science 256, 1405–1411. Planetary Science Letters 168, 79–99.
Leach, D.L., Bradley, D.C., Pisarevsky, S.A., Taylor, R.D., Gardoll, S.J., 2010. The genesis of Whitmeyer, S.J., Karlstrom, K.E., 2007. Tectonic model for the Proterozoic growth of
sediment-hosted lead-zinc deposits in Earth history. Economic Geology 105, 593–625. North America. Geosphere 3, 220–259. doi:10.1130/GES00055.
Li, Z.-X., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., Williams, H., Hoffman, P.H., Lewry, J.F., Monger, J.W.H., Rivers, T., 1991. Anatomy of
Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S., Natapov, L.M., Pease, V., North America: thematic geologic portrayals of the continents. Tectonophysics 187,
Pisarevsky, S.A., Thrane, K., Vernikovsky, V., 2008. Assembly, configuration, and break- 117–134.
up history of Rodinia: a synthesis. Precambrian Research 160, 179–210. Wilson, J.T., 1968. Static or mobile Earth: the current scientific revolution. Proceedings
Meyer, C., 1981. Ore-forming processes in geologic history of the earth. Economic of the American Philosophical Society 112, 309–320.
Geology, pp. 6–41. 75th Anniversary Volume. Woolley, A.R., Kjarsgaard, B.A., 2008. Carbonatite occurrences of the world: map and
Meyer, C., 1988. Ore deposits as guides to geologic history of the earth. Annual Reviews database. Geological Survey of Canada Open File 5796, 1 CD-ROM + 1 map.
of Earth Science 16, 147–171. Worsley, T.R., Nance, D., Moody, J.B., 1984. Global tectonics and eustasy for the past 2
Moores, E.M., 1991. Southwest US-East Antarctic (SWEAT) connection: a hypothesis. billion years. Marine Geology 58, 373–400.
Geology 19, 425–428. Worsley, T.R., Nance, R.D., Moody, J.B., 1986. Tectonic cycles and the history of the
Moores, E.M., 2002. Pre–1 Ga (pre-Rodinian) ophiolites: their tectonic and environ- earth's biogeochemical and paleoceanographic record. Paleoceanography 1,
mental implications. GSA Bulletin 114, 80–95. 233–263.
Murphy, J.B., Nance, R.D., 2003. Do supercontinents introvert or extrovert? Sm-Nd Zhao, G., Cawood, P.A., Wilde, S.A., Sun, M., 2002. Review of global 2.1–1.8 Ga orogens:
isotope evidence. Geology 31, 873–876. implications for a pre-Rodinia supercontinent. Earth Science Reviews 59, 125–162.
 Supplementary Data For:

Secular trends in the geologic record

and the supercontinent cycle1
Dwight C. Bradley

U.S. Geological Survey, 4200 University Drive, Anchorage, Alaska 99508 U.S.A.
dbradley@usgs.gov

Additional discussion of secular trends plotted in Figures 3-45.

This electronic-only supplement provides additional information about the secular trends featured in the
main body of the paper, with particular emphasis on secular trends that are new to this paper. Compilations
covering the whole Earth and all of geologic time are not easy. If broader inferences are to be based these
secular trends, it is important that their strengths and weaknesses be acknowledged.

Indices of supercontinentality (supplement to Section 4.1)

The aim of Figures 3 and 4 is to quantify what is plainly seen in a sequence of Phanerozoic
paleogeographic maps: that the continents were dispersed, then gathered into a single supercontinent, and then
dispersed once again. In Figure 3, the y axis is shown in square kilometers and in percentage of total continental
area, using 214 million km sq for the total (from Cogley, 1984).

Passive margins (supplement to Sections 4.2 and 6.2)

Figures 11, 12, 13, 27, and 28 are from my recent synthesis of passive margins through time (Bradley,
2008). A lengthy online appendix included writeups of each passive margin that gave reasons for the tectonic
interpretation, and for the ages picked for the start date and end date (rift-drift transition and onset of collision).
The dataset is a distillation of the tectonic interpretations of >80 passive margins in orogenic belts.
Undoubtedly, mistaken interpretations were included—such is the difficulty of unraveling orogenic histories—
but the overall form of the passive margin age distribution (Fig. 11) is probably robust. Each passive margin
was assigned a quality ranking from A to D, based on my confidence in the tectonic interpretation and my
perception of the reliability of the age control. Passive margins of each quality group have similar age
distributions. In the passive-margin record, the interval from ca. 1650 to ca. 1050 Ma is problematic because the
tectonic interpretations are debatable and the age controls are poor on the few putative margins of that age. The
plot of passive-margin lifespans (Fig. 27) has greater potential for error (Bradley, 2008). Figure 28 is new and
derived from the lifespan data. By analogy with the modern passive margins, each ancient passive margin is
assumed to have been flanked by seafloor that, by definition, has the same age as the "start date" of the margin.
Thus the end date minus the start date gives the age of that strip of oceanic crust at the time that it was
subducted, just before the start of collision. In Figure 27, the x-value is the start date, and the y-value is the
lifespan.

Orogenic granites and detrital zircons (supplement to Sections 4.3 and 6.9)

The plot of zircon ages from orogenic granites in Fig. 14 (from appendix in Condie et al., 2009a) is the
most recent iteration of a series of such plots by Condie. Some reasons for the paucity of Phanerozoic ages were
discussed in Section 4.2. Another reason is that the 40Ar/39Ar method is widely used instead of the U-Pb method
for Mesozoic and younger igneous rocks, so zircon ages are not reported. In an earlier version (Fig. A19),
Condie (1997) used (but did not fully describe) a weighting procedure designed not to over-represent well-
studied igneous belts. Condie's plots improve on earlier plots of the same type by Gastil (1960), Hurley and
Rand (1969), and Glikson (1983; Fig. A20). Those early plots were largely based on K/Ar and Rb/Sr data that

1
Bradley, D.C., 2011. Secular trends in the geologic record and the supercontinent cycle. Earth-
Science Reviews. DOI: 0.1016/j.earscirev.2011.05.003
would not meet today's standards of precision or accuracy. In addition, many more age determinations are now
available, and this is especially important because remote parts of the world were poorly represented in the
earlier tallies. Despite these differences in methodology, the major peaks in Figure 14 were already evident by
1960.
Two plots are shown of detrital zircons from modern river sands (Figs. 15 and A21). The advantage of this
method is in capturing zircon populations from broad regions in a small number of samples. Potential
shortcomings of the detrital record include (1) destruction of high-U zircons in transit from bedrock source to
river delta, (2) under-representation of small zircons, (3) over-representation of plutonic sources that are
anomalously rich in zircon, (4) over-representation of zircons from areas of rapid erosion, and (5) sampling bias
against young orogens and magmatic arcs. A study designed to incorporate and properly weight data from
smaller rivers would be useful.
Figure 16 is a new plot based on my own compilation of detrital zircons of the world. It is far from
complete and could be much improved by devising a way to guard against undue influence of data from heavily
sampled regions. In Figure 40, the >26,000 igneous ages have been plotted as a function of the depositional age
of the host sandstone or metasandstone. Each data point that plots below the diagonal line is in the forbidden
zone: either the true depositional age is younger than shown, the true zircon age is older, or both.
(Note added in proof: Voice et al. (2011) published a new compilation of about 200,000 detrital zircon ages
after final version of the present paper was submitted. The maxima and minima are similar to those shown in
Figures 14-16.)

Metamorphic rocks (supplement to Sections 4.6 and 6.4)

The histograms showing occurrences of blueschist, granulite-, and eclogite-facies metamorphism (Figs. 20,
21, and 30) were constructed from data tables in Brown (2007). The plots of metamorphic pressures and
geothermal gradient (Figs. 31 and 32) were redrafted from Brown (2008). Brown (2008) argued that the coeval
existence of two distinctive P-T regimes as far back as the Neoarchean is an earlier manifestation of the classic
paired metamorphic belts of the Phanerozoic, and constitutes evidence for plate tectonics.

Carbonatites (supplement to Section 4.7)

The histogram in Figure 22 was plotted from an Excel table in Woolley and Kjarsgaard (2008). The
published histogram in that paper was binned in 75-m.y. increments. Only about half of the carbonatites listed
in Woolley and Kjarsgaard's comprehensive study were assigned an age.

Large Igneous Provinces (supplement to Section 4.8)

The age distribution of ancient LIP s (Fig. 23) is based on occurrences of flood basalts, layered mafic
intrusions, mafic dike swarms, and komatiites. A komatiite need not be regionally extensive to count; that is, the
"large" in Large Igneous Province is not a prerequisite for this census. An arc origin for any of the komatiites
(e.g., Grove and Parman, 2004) would exaggerate the LIP record before about 1800 Ma.

Oceanic crust and ophiolites (supplement to Sections 6.2 and 6.3)

Figure 8 was adapted from Sclater et al. (1980). The original plot employed bins of somewhat uneven
duration corresponding to the subdivisions of the Mesozoic-Cenozoic time scale as it stood in 1980; the bins
have been resized in the x-dimension to conform to the Gradstein and Ogg (2004) time scale.
Figure 29 was pieced together from compilations by Dilek (2003) for ophiolites between 0 and 1000 Ma,
and from Moores (2002) for ophiolites from 1000 to ca. 3500 Ma. Moores (2002) provided a summary data
table, whereas Dilek did not, making it harder to merge the two datasets. Dilek (2003) plotted the age
distribution of ophiolites in histogram form but with uneven bin sizes: in the Phanerozoic, bins appear to
correspond to various uneven divisions of a geologic time scale that predates Gradstein and Ogg (2004),
perhaps 5 to 15 m.y. in duration, whereas bins in the Neoproterozoic appear to be 20 m.y. wide. To more
faithfully portray the merged data, everything in Figure 29 has been replotted in 50 m.y. bins. A few
Phanerozoic and Neoproterozoic ophiolites may have been inadvertently assigned to the next-adjacent bin in the
process. In those cases where Moores (2002) gave an age range, I used the midpoint.
Mantle-derived igneous rocks (supplement to Section 6.5)

Plots of density, mineralogy, and chemistry of average subcontinental lithospheric mantle through time
(Figs. 35, A27, and A28) were derived from data tables in Poudjom Djomani et al. (2001) and Griffin et al.
(2003). The xenoliths analyzed in these studies were assumed to be about the same age as the upper crust into
which they were emplaced. Age groupings are so broad that only generalized, first-order trends can be detected.
A schematic plot of mantle redox conditions by Delano (2001) is shown in Figure A35. The mantle's
oxidation state was inferred from whole-rock abundances of Cr and V in ancient volcanics, and from the
composition of Cr-rich spinels in ancient volcanics. This study was based on samples whose ages were only
broadly assigned (e.g. Archean, Proterozoic, etc.). The oxidation state of the mantle has remained constant since
the Eoarchean. The suggestion that oxidation of the atmosphere at around the Archean-Proterozoic boundary
was due to a mantle overturn event (Kump et al., 2001) is therefore untenable.

Massif anorthosites (supplement to Section 6.6)

The histogram in Figure 27 was constructed from tabulated data in Ashwal (1993). Anorthosites having age
assignments with errors >50 m.y. were excluded. Given the worldwide advancement of zircon geochronology
since the early 1990s, it would be worth updating this plot.

Sedimentary recycling (supplement to Section 6.9)

Each sedimentary rock analyzed is a mix of clastic grains from various sources, so each data point
represents an average, at a given time, of crustal residence ages from the sediment catchment area. The red line
is x = y and below it is the forbidden zone. Hafnium isotopes from sediments show essentially the same age
distribution as neodymium isotopes (Fig. A63). The new plot of igneous age versus depositional age (Fig. 40)
is based on the global compilation of >26,000 detrital zircons shown in Figure 16. The clustering of data in
Figures 39 and 40 is probably due to the disproportionate abundance of strata dating from times of
supercontinent disassembly and assembly.

Sea level (supplement to Section 6.12)

An alternative to Hallam's (1992) sea level curve has been produced by Haq et al. (1987) for the latter half
of the Phanerozoic, and Haq and Schuller (2008) for the earlier half. The curves are broadly in agreement but
differ in some details.

References cited in Appendix only

Abbott, D.H., and Isley, A.E., 2002. Extraterrestrial influences on mantle plume activity. Earth and Planetary
Science Letters 205, 53-62.

Artemieva, I.M., and Mooney, W.D., 2001. Thermal thickness and evolution of Precambrian lithosphere: A
global study. Journal of Geophysical Research 106, 16387-16414.

Artemieva, I.M., and Mooney, W.D., 2002. On the relations between cratonic lithosphere thickness, plate
motions, and basal drag. Tectonophysics 358, 211–231.

Bekker, A., Slack, J.F., Planavsky, N., Krapež , B., Hofmann, A., Konhauser, K.O., and Rouxel, O.J., 2010.
Iron formation: tthe sedimentary product of a complex interplay among mantle, tectonic, oceanic, and
biospheric process. Economic Geology 105, 467-508.

Bois, C., Bouche, P., and Pelet, R., 1982. Global geologic history and distribution of hydrocarbon reserves.
American Association of Petroleum Geologists Bulletin 66, 1248-1270.

Condie, K.C., 2003b. Incompatible element ratios in oceanic basalts and komatiites: Tracking deep mantle
sources and continental growth rates with time. Geochemistry, Geophysics, Geosystems 4, 1-28.
doi:10.1029/2002GC000333.
Condie, K.C., 2008. Did the character of subduction change at the end of the Archean? Constraints from
convergent-margin granitoids. Geology 36, 611-614.

Condie, K.C., Des Marais, D.J., and Abbott, D., 2001. Precambrian superplumes and supercontinents: a record
in black shales, carbon isotopes, and paleoclimates? Precambrian Research 106, 239–260.

de Wit, M.J., and Ashwal, L.D., 1997, Convergence towards divergent models of greenstone belts, in M.J. de
Wit and L.D. Ashwal, editors, Greenstone Belts: Oxford University Press, New York, p. ix-xix.

Delano, J.W., 2001. Redox history of the earth's interior since ~3900 Ma: Implications for prebiotic molecules.
Origins of life and evolution of the Biosphere 31, 311-341.

Dunlap, W.J., 2000. Nature’s diffusion experiment: The cooling-rate cooling-age correlation. Geology 28, 139–
142.

Durrheim, R. J., and Mooney, W.D., 1994. Evolution of the Precambrian Lithosphere: Seismological and
geochemical constraints. Journal of Geophysical Research 99, 15359-15374.

Evans, D.A.D., 2006. Proterozoic low orbital obliquity and axial-dipolar geomagnetic field from evaporite
palaeolatitudes. Nature, 444, 51-55.

Gastil, G., 1960. The distribution of mineral dates in space and time. American Journal of Science 258, 1-35.

Glikson, A.Y., 1983. Geochemical, isotopic, and paleomagnetic tests of early sial-sima patterns; the
Precambrian crustal enigma revisited, in Medaris, L.G., Jr., Byers, C.W., Mickelson, D.M., and Shanks,
W.C., editors, Proterozoic geology; selected papers from an international Proterozoic symposium. Geological
Society of America Memoir 161, 95-117.

Gough, D.O., 1981. Solar interior structure and luminosity variations. Solar Physics 74, 21-34.

Hartmann, W.K., Ryder, G., Dones, L., and Grinspoon, D., 2000. The time-dependent intense bombardment of
the primordial Earth/Moon system, in R.M. Canup and K. Righter, editors, Origin of the Earth and Moon.
Tucson, University of Arizona Press, 493-512.

Hazen, R.M., Papineau, D., Bleeker, W., Downs, R.T., Ferry, J.M., McCoy, T.M., Sverjensky, D.A., and Yang,
H., 2008. Mineral evolution. American Mineralogist, 93, 1693–1720.

Holland, H. D., 2006. The oxygenation of the atmosphere and oceans. Philosophical Transactions of the Royal
Society B, 361, 903–915.

Jewell, P.W., 2000, Bedded barite in the geological record. SEPM Special Publication 66, 147-161.

Johnson, C.M., Beard, B.L., and Roden, E.E., 2008. The iron isotope fingerprints of redox and biogeochemical
cycling in modern and ancient Earth. Annual Review of Earth and Planetary Sciences 36, 457-493. doi:
10.1146/annurev.earth.36.031207.124139.

Komiya, T., 2007. Material circulation through time; chemical differentiation within the mantle and secular
variation of temperature and composition of the mantle, In: Yuen, D.A., Maruyama, S., Karato, S.-I., and
Windley, B.F., (Eds.), Superplumes: beyond plate tectonics. Springer, Dordrecht, Netherlands, pp. 187-234.

Kump, L.R., Kasting, J.F., and Barley, M.E., 2001. Rise of atmospheric oxygen and the ''upside-down'' Archean
mantle: Geochemistry Geophysics Geosystems 2, Paper number 2000GC000114. ISSN: 1525-2027.

Leach, D.L., Bradley, D.C., Pisarevsky, S.A., Taylor, R.D., and Gardoll, S.J., 2010. The genesis of sediment-
hosted lead-zinc deposits in Earth history. Economic Geology 105, 593-625.
Lowenstein, T.K., Hardie, L.A, Timofeeff, M.N., and Demicco, R.V., 2003. Secular variation in seawater
chemistry and the origin of calcium chloride brines: Geology 31, 857-860.

McLennan, Taylor, S.R., and Hemming, S.R., 2005. Composition, differentiation, and evolution of continental
crust: constraints from sedimentary rocks and heat flow, in M. Brown and T. Rushmer, editors, Evolution and
Differentiation of the Continental Crust, Cambridge University Press, 92-134.

Muller, R. A., 2002. Measurement of the lunar impact record for the past 3.5 b.y. and implications for the
Nemesis theory. Geological Society of America Special Paper 356, 659-665.

Orem, W.H., and Finkelman, R.B., 2007. Coal Formation and Geochemistry. Treatise of Geochemistry 7, 191-
222.

Retallack, G.J., 2007. Coevolution of life and Earth. Treatise of Geophysics 9, 295-320.

Rey, P.F., and Coultice, N., 2008. Neoarchean lithospheric strengthening and the coupling of Earth’s
geochemical reservoirs. Geology 36, 635-638.

Rohde, R.A., and Muller, R.A., 2005. Cycles in fossil diversity. Nature 434, 208-210.

Shaviv, N.J., 2003. The spiral structure of the Milky Way, cosmic rays, and ice age epochs on Earth. New
Astronomy 8, 39–77.

Sundquist, E.T., and Visser, K., 2003. The geologic history of the carbon cycle: Treatise on Geochemistry 8,
425-472.

Taylor, S. R. and McLennan, S. M., 1985. The continental crust: its composition and evolution. Oxford,
Blackwell Scientific Publications, 312 p.

Tkachev, A.V., 2011. Evolution of metallogeny of granitic pegmatites associated with orogens throughout
geological time. Geological Society, London, Special Publications, 350, 7–23

Valley, J.W., Lackey, J.S., Cavosie, A.J., Clechenko, C.C., Spicuzza, M.J., Basei, M.A.S., Bindeman, I.N.,
Ferreira, V.P., Sial, A.N., King, E.M., Peck, W.H., Sinha, A.K., and Wei, C.S., 2005. 4.4 billion years of
crustal maturation: oxygen isotope ratios of magmatic zircon. Contributions to Mineralogy and Petrology
150, 561–580. DOI 10.1007/s00410-005-0025-8.

Voice, P.R., Kowalewski, M., and Eriksson, K.A., 2011. Quantifying the timing and rate of crustal evolution:
global compilation of radiometrically dated detrital zircon grains. Journal of Geology 119, 109–126. DOI:
10.1086/658295

Figures A1-A86

Figures A1-A86 show additional plots of secular trends that were compiled for this study and redrafted at
the same scales as Figures 3-44. A few of these are redundant and a few others are of historical interest only—
but most are in the Appendix because they bear only peripherally, if at all, on the supercontinent cycle. Table
A1 summarizes the contents of Figures A1-A86 and takes the place of as many figure captions.
 Ceno- Mesozoic Paleozoic Ceno- Mesozoic Paleozoic Ceno- Mesozoic Paleozoic
zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed. zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed. zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed.
100
90 Volcanogenic massive sulfide deposits A1 A5 NaCl and CaSO4 A9

106
Mineralogy of

NaCl, CaSO4 vol (km3)

Number of deposits (Slack et al., 2010) reef limestones in marine evaporites
75 75
(Veizer & Mackenzie, 2003) (Holser, 1984)

Percent
60
50 aragonite CaSO4
45

105
30
25 NaCl
15
calcite

104
0 0
0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500

100 8000
14
A2 aragonite A6 A10
Number of occurrences

12 UHP Present-day halite mass

75 6000 (Hay et al., 2006)
10

Mass (1015 g)
calcite

Percent
8 BS
50 4000
6
Mineralogy of
4 oolitic limestones
Age distribution of bluesch st and 25 2000
2 ultra-high-pressure metamorphic rocks (Morse and Mackenzie, 1990)
(Brown 2007)
0 0 0

0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500

25 14000
20
A3 Age distribution of oolitic limestone A7 A11
Number of occurrences

dolomites
12000
Age distribution of 20 Reconstructed original
Sediment mass (1016 tons)

limestones (Wilkinson et al., 1985)

Mass (1015 g)
15 sedimentary carbonates 10000 halite mass
(Mackenzie & Morse, 1992) 15 (Hay et al., 2006)
8000
10 6000
10

4000
5
5
2000

0 0 0
0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500

14 60
Calcite/dolomite mass ratio

A8 A12

Percent of world reserves

12 Calcite to dolomite ratio 50 Age distribution of 30 Age distribution
Number of deposits

in carbonate rocks oolitic marine ironstones

10 of coal (Orem &
40 (Bekker et al., 2010)
(Mackenzie Finkelman, 2007)
8 & Morse, 1992) 20
30
6
20
4 10

2 10
A4
0 0 0

0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500
Age (Ma) Age (Ma) Age (Ma)

Bradley, Figures A1-A12

 Ceno- Mesozoic Paleozoic Ceno- Mesozoic Paleozoic
zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed. zoic Cret. Jur. Trias. Perm. P. Miss. Dev. Sil. Ord. Cam. Ed.
1000
5
Reserves (106 m3/m.y.) A13 A17
800
Age distribution of
oil source rocks -5
600

δ18O
(Bois et al., 1982)
400
-15

200 Oxygen isotopic composition of seawater

(Shields & Veizer, 2002)
0 -25
0 100 200 300 400 500 0 100 200 300 400 500

A14
40 CaCl2 seas CaCl2 versus
Sulfur isotopic
MgSO4 seas
50
composition of seawater
A18
(Lowenstein
mmol/kg H2O

30 40
(Shields & Veizer, 2002)
et al., 2004)

δ34S
20 30

20
10
MgSO4
MgSO4 MgSO4 seas seas 10
seas
0
0 100 200 300 400 500 0 100 200 300 400 500
Number of short-lived genera
1000

1500
Sr concentration in biological
A15 A19
low magnesium calcite 800 Fossil diversity
(Veizer & Mackenzie, 2003)
Neogene not pltoted

(Rohde & Muller, 2005)

1000 600
Sr ppm

400
500
200

0 0

0 100 200 300 400 500 0 100 200 300 400 500

0
15 A16 Minimum latitude A20
of continental glaciation
Latitude (degrees)

(Sundquist & Visser, 2003)

5 30
δ13C

-5

60
-15
Carbon isotopic composition of seawater
(Shields & Veizer, 2002)
-25 90

0 100 200 300 400 500 0 100 200 300 400 500
Age (Ma)

Bradley, Figures A13-A20

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean

Age distribution of A21 Age distribution of A25 10 Thickness of continental crust A29
12

Depth to Moho (km)

orogenic granite greenstone belts (Durrheim & Mooney, 1994)
20
(Condie, 1997) (Condie, 1994)
Percent area
10
30

Number
8 Archean cratons
40
6
50
all cratons
4
60
2 70
0 80
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

100 80
Ol
Rb-Sr ages of granites A22 A26 70 A30
80 (Glikson, 1983) 2000 Age distribution of

Modal percentage
volcanic-hosted massive sulfide 60 Mineralogy of SCLM

Tonnage (Mt)
deposits (Huston et al., 2010) (Griffin et al., 2003)
Number

60 50

40 “Archons”
40 1000 “Tectons” “Protons”
30
Opx
n = 612 20
20
10 Gt
0
Cpx
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

10
Major element chemistry of SCLM
20 1044 1912
Detrital zircons from A23 12 A27 (Poudjom Djomani et al., 2001) A31
Juvenile crust
modern river sands (Condie, 2005) 8
10

Weight percent
15 (Rino et al., 2008)
FeO
Percent area

8 6
Number

574 2691
SiO2 / 10
10 6
4 MgO / 10
4
5 Al2O3
2
2
CaO
0 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

70 0

Lithospheric thickness (km)

Pegmatites A24 A28 A32
Cumulative area (percent)

525
60 975 F
100
Number of pegmatite fields

275
(Tkachev, 2011)
1875 100
50 AM

80 B
2625
40 C
60 M&B R&S 200
30
V&J
O'N
40
20 D&W
300 Thickness of continental
2075
2825 Continental
10 1175 20 growth curves H&R
M&T lithosphere
AL
(Komiya, 2007) (Artemieva & Mooney, 2002)
400
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A21-A32

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
0.750 40

30 Thickness of oceanic crust A33 A37 MgO in komatiites A41

Initial 87Sr/86Sr
(Moores, 2002) 0.740
Initial 87Sr/86 (Condie and O’Neill, 2011)

Average MgO
Thickness (km)

in granites 30
0.730
20 (Glikson, 1983)
0.720
20
10
0.710

0 0.700 10
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

14
70
Oxygen isotopic A34 A38 A42
δ18O in zircon

12
compositions of zircon 60 SiO2

oxidizning
10 (Valley et al., 2005) oxidized mantle

Oxidation State

Oxide wt pct.
mantle 50
redox
8 Major elements in avg. continental crust
unknown 40
Mantle redox state (Condie, 1993)
6
(Delano, 2001) 30

reducing
reduced mantle
4 at time of 20 Al2O3
core formation
2 10
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

20 60 6
Reduced heat flow, mW/m2

Hafnium isotopic compositions

of zircons (Hawkesworth et al., 2010)
A35 A39 FeOT A43
50 5
10

Oxide wt pct.
40 4 CaO
εHf (t)

Na2O
0 CHUR 30 3 K 2O

20 2 MgO
-10
Continental heat flow Major elements in avg. continental crust
10 1
(Artemieva & Mooney, 2001) (Condie, 1993)
-20 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

20 20
Hafnium isotopes
A36 A40 A44

Abundance (ppm)
in mantle-derived rocks Percent komatiite 16 Pb
1 15
dm (Bennett, 2003) Trace elements in avg. cont. crust
of total volcanics
(Condie, 1993)
12
Percent

(Condie and O’Neill, 2011)

εHf

1 10
8 Th
5
4 U
pm
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A33-A44

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
1000
20
A45 MgO in granitoids A49 A53

Cooling rate (max °C/m.y.)

N-MORB
Nb/Th in basalts & granites 3
(Condie 2003, 2008) (Condie, 2008) Orogenic cooling rate
Basalt

Weight percent
15 Ophiolite TTG
100 (Dunlap, 2000)
Komatiite CA
2
Nb/Th

TTG
CA granite
10
Pr mitive mantle

1
10
5

0 0 1
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

2
80 A46 Europium anomaly A54
Nb/U in basalts & granites
in granitoids

Plateau elevation (km)

basalt 10
1.5 Syncollisional elevation
(Condie 2003, 2008) komatiite
60
TTG 8 Rey et al. (2008)

Eu/Eu*
CA granite
Nb/U

1.0 6
40
Primitive mantle
4
20 0.5
2
0 (Condie, 2008) A50
0 0 0
1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

10 100
Th/U in basalts & granites Thorium/uranium La/Yb in granitoids A51 16 A55
La/Yb nornmalized

basalt
8 Condie 2003, 2008 komatiite
TTG
CA granite

Th (ppm)
6 12
Th/U

4 Prim t ve mant e 10
8 Thorium in
sedimentary rocks
2 (McLennan et al., 2005)
A47 (Condie, 2008) 4
0
0 1000 2000 3000 4000 0 1000 2000 3000 4000

2.0 1.5
K2O/Na2O in granitoids
(Condie, 2008)
A48 100 Sr/Y in granitoids
A52 A56
1.5 CA
(Condie, 2008)
K2O/Na2O

1.0
TTG
Sr/Y

Th/Sc
MCC
1.0
10 0.5
Th/Sc in
0.5
sedimentary rocks
(McLennan et al., 2005)
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A45 -A56

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
6

Chemical Index of Alteration

6 100

A57 A61 5
A65
90

No Phanerozoic data
5
4

Number
80 Age distribution of evapor tes
3 (Evans, 2006)
4
Th/U

70
2

Th/U in sedimentary rocks 60 1

3 Chemical Index of Alteration in shales
(McLennan et al., 2005) (Condie et al., 2001)
50 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

1.2 8

A58 1.0 A62 A66

Black shale/total shale

No Phanerozoic data
Europium anomaly Age distribution of
1.0 6
in sedimentary rocks 0.8 sedimentary barite deposits
(Taylor & McLennan, 1985) (Jewell, 2000)

Number
Eu/Eu*

0.8 0.6 4

0.4 Black shale/total shale

0.6 (Condie et al., 2001) 2
0.2

0.4 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

100
16
A59 4 mud A63 A67
Mantle sep. age (Ga)

sandstone
14 80

12 3 Clastic-dominated

Pb+Zn (Mt)
60
lead-zinc deposits
ΣREE/ΣLREE

10
2 (Leach et al., 2010)
40
8 Hafnium isotopes
REE in sedimentary rocks 1 in sedimentary rocks
20
6 (Taylor & McLennan, 1985) (Vervoort et al., 1999)
4 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

5 40
La/Sc in sedimentary rocks
5
(Taylor & McLennan, 1985) A60 A64 35
Bauxite deposits A68
4 Age distribution of carbonate rocks
4 30 (Retallack, 2010)
(Ronov et al., 1991)

Number
1020g/m.y.

25
3
3
La/Sc

20
2 15
2
metamorphosed
10
1 1 unmetamorphosed
5
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A57-A68

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
100,000 5
Oxygen isotopic composition of seawater
Sedimentary manganese deposits A69 (Shields & Veizer, 2002) A73 1 A77
(Maynard, 2010)

δ180PDB in calcite
10,000
low-C, -S
Manganese (Mt)

0
clastic rocks
-5

δ56Fe (0/00)
1,000
Volcanic-hosted -1

Sediment-hosted Iron isotopes in shale

100
-2 (Johnson et al., 2008)
-15

10 -3

n = 9557
1 -25 -4

0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

250 0.0005 0.4

K C
Sedimentary-hosted A70 A74
From paleosols
A78
200 copper deposits From ironstones
0.0004 Ni/Fe in banded iron formation 0.3

O2 (atmospheres)
(Hitzman et al., 2010) From δ13C
Copper (Mt)

(Konhauser et al., 2009) Mass balance model

Atmospheric oxygen

Molar Ni / Fe
150 K—Kupferschiefer (Retallack, 2007)
0.0003 0.2
C—Central African Cu Belt
W—White Pine
100 U—Udokan 0.0002
red curve—mineralization age 0.1
brown boxes—host-rock ages
50 0.0001
U 0
W ?
0 0.0000
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

14 5 500

12 Uranium resources A71 A75 Oxygen content of the shallow oceans

A79
Europium anomaly 400
(Cuney, 2010) (Holland, 2006)
10
Others 4 in iron formation Algoma-type
Uranium (Mt)

[Eu/Eu*]SN

Phosphorites (Huston et al., 2004) BIFs

PO (µmol)
8
300
Black shales
3
6
Iron oxide-copper-gold
200

2
Quartz pebble conglomerate
4
2
Superior-type 100
2
BIFs
0 1 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

500
Carbon isotopic
15 n = 15797
composition A72 1 A76 A80
400
δ13C in calcite

of seawater Oxygen content of the deep oceans

0

PO (µmol)
low-C, -S clastic rocks
5 (Holland, 2006)
δ56Fe (0/00)

300
-1

2
-5
Iron isotopes 200
(Shields & Veizer, -2 in BIF and pyrite
-15
2002) (Johnson et al., 2008) pyrite
-3 100
open symbol = poor age control BIF
closed symbol = better age control
-25 -4 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A69-A80

 Phanero- Proterozoic Archean Phanero- Proterozoic Archean Phanero- Proterozoic Archean
zoic Hadean zoic Hadean zoic Hadean
5000 3

Impact rate relative to today

Cumulative number
A81 10 9
A85
4000
Moon-forming
collision
2
3000 106
Terrestrial impact flux Meteor cosmic exposure ages

N / bin
(Hartmann et al., 2000) (Shaviv, 2003)
2000
103
Number of mineral species 1
1000
(Hazen et al., 2008)
1 A83
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000

Solar luminosity (past/present)

0.2 1.0
A82

Spherules per 50 m.y. bin

7
Lunar impacts A84 A86
0.8
6 from spherule ages
Relative probability

Large Earth impacts

(Abbott & Isley, 2002) 5 (Muller, 2002)
0.6
4 Solar luminosity
0.1
(Gough, 1981)
3 0.4

2
0.2
1

0.0 0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
Age (Ma) Age (Ma) Age (Ma)

Bradley, Appendix Figures A81-A86

Table A1. Summaries of time-series plots in Figures A1-A86.
Fig. Variable Time Reference and comments
no. span
(Ma)
A1 Age distribution of 0-550 Bekker et al. (2010)
volcanic-hosted massive
sulfide deposits
A2 Age distribution of 0-550 Brown (2007)
blueschist and ultra-
high-pressure
metamorphic rocks
A3 Age distribution of 0-550 Mackenzie and Morse (1992)
sedimentary carbonates
A4 Calcite to dolomite ratio 0-550 Mackenzie and Morse (1992)
in sedimentary rocks
A5 Mineralogy of reef 0-550 Veizer and Mackenzie (2003)
limestones
A6 Mineralogy of oolitic 0-550 Morse and Mackenzie (1990)
limestones
A7 Age distribution of 0-550 Wilkinson et al. (1985)
oolitic limestone
A8 Age distribution of 0-550 Bekker et al. (2010)
oolitic marine
ironstones
A9 Age distribution of 0-550 Holser (1984)
evaporites
A10 Present day halite mass 0-550 Hay et al. (2006)
A11 Reconstructed halite 0-550 Hay et al. (2006)
mass
A12 Age distribution of coal 0-550 Orem and Finkelman (2007)
A13 Age distribution of oil 0-550 Bois et al. (1982)
source rocks
A14 Phanerozoic seawater 0-550 Lowenstein et al. (2004)
chemistry MgSO4 and
CaCl2
A15 Strontium 0-550 Veizer and Mackenzie (2003).
concentrations in
biological low
magnesium calcite
A16 Carbon isotopic 0-550 Shields and Veizer (2002)
composition of seawater
A17 Oxygen isotopic 0-550 Shields and Veizer (2002)
composition of seawater
A18 Sulfur isotopic 0-550 Shields and Veizer (2002). Gray boxes
composition of seawater represent data from marine evaporites.
A19 Fossil diversity 0-550 Rohde and Muller (2005)
A20 Minimum latitude of 0-550 Sundquist and Visser (2003)
continental glaciation
A21 Age distribution of 0-4560 Condie (1997)
orogenic granites (old
version)
A22 Rb-Sr ages of granites 0-4560 Glikson (1983)
A23 Detrital zircons from 0-4560 Rino et al. (2008)
modern river sands
A24 Pegmatites 0-4560 Tkachev (2011)
A25 Age distribution of 0-4560 Condie (1994)
greenstone belts
A26 Age distribution of 0-4560 Huston et al. (2010)
volcanic-hosted massive
sulfide deposits
A27 Juvenile crust 0-4560 Condie (2005)
A28 Continental growth 0-4560 Komiya (2007); see caption to his Figure
curves 14 for references
A29 Thickness of continental 0-4560 Durrheim & Mooney (1994)
crust
A30 Mineralogy of 0-4560 Griffin et al. (2003)
subcontinental
lithospheric mantle
A31 Major elements in 0-4560 Poudjom Djomani et al (2001)
subcontinental
lithospheric mantle
A32 Thickness of continental 0-4560 Artemieva and Mooney (2002)
lithosphere
A33 Thickness of oceanic 0-4560 Moores (2003)
crust
A34 Oxygen isotopic 0-4560 Valley et al. (2005)
compositions of zircons
A35 Hafnium isotopic 0-4560 Hawkesworth et al. (2010). The green curve
compositions of zircons represents depleted mantle.
A36 Hafnium isotopes in 0-4560 Bennett (2003)
mantle-derived rocks
A37 Initial 87Sr/86Sr in 0-4560 Glikson (1983)
granites
A38 Mantle redox state 0-4560 Delano (2001)
A39 Continental heat flow 0-4560 Artemieva and Mooney (2001)
A40 Percent komatiite of 0-4560 Condie and O’Neill (2011), after de Wit
total volcanics and Ashwal (1997)
A41 MgO in komatiites 0-4560 Condie and O’Neill (2011)
A42 Major elements in avg. 0-4560 Condie (1993)
continental crust 1
A43 Major elements in avg. 0-4560 Condie (1993)
continental crust 2
A44 Trace elements in avg. 0-4560 Condie (1993)
continental crust
A45 Nb/Th in basalts and 0-4560 Condie (2003b, 2008)
granites
A46 Nb/U in basalts and 0-4560 Condie (2003, 2008)
granites
A47 Th/U in basalts and 0-4560 Condie (2003, 2008)
granites
A48 K2O/Na2O in granitoids 0-4560 Condie (2008)
A49 MgO in granitoids 0-4560 Condie (2008)
A50 Europium anomaly in 0-4560 Condie (2008)
granitoids
A51 La/Yb in granitoids 0-4560 Condie (2008)
A52 Sc/Y in granitoids 0-4560 Condie (2008)
A53 Orogenic cooling rate 0-4560 Dunlap (2000)
A54 Syncollisional elevation 0-4560 Rey and Coultice (2008)
A55 Thorium in sedimentary 0-4560 McLennan et al. (2005)
rocks
A56 Th/Sc in sedimentary 0-4560 McLennan et al. (2005)
rocks
A57 Th/U in sedimentary 0-4560 McLennan et al. (2005)
rocks
A58 Europium anomaly in 0-4560 Taylor and McLennan (1985)
sedimentary rocks
A59 REE in sedimentary 0-4560 Taylor and McLennan (1985)
rocks
A60 La/Sc in sedimentary 0-4560 Taylor and McLennan (1985)
rocks
A61 Chemical Index of 0-4560 Condie et al. (2001)
Alteration in shales
A62 Black shale/total shale 0-4560 Condie et al. (2001)
A63 Hafnium isotopes in 0-4560 Vervoort et al. (1999)
sedimentary rocks
A64 Age distribution of 0-4560 Ronov et al. (1991)
carbonate rocks
A65 Age distribution of 0-4560 Evans (2006)
evaporites
A66 Age distribution of 0-4560 Jewell (2000)
sedimentary barite
deposits
A67 Clastic dominated lead- 0-4560 Leach et al. (2010)
zinc deposits
A68 Bauxite deposits 0-4560 Retallack (2010), who presented a similar
plot for laterites, along with detailed time
series for 0-300 Ma for both
A69 Sedimentary manganese 0-4560 Maynard (2010)
deposits
A70 Sediment-hosted copper 0-4560 Hitzman et al. (2010)
deposits
A71 Uranium resources 0-4560 Cuney (2010)
A72 Carbon isotopic 0-4560 Shields and Veizer (2002)
composition of seawater
A73 Oxygen isotopic 0-4560 Shields and Veizer (2002)
composition of seawater
A74 Ni/Fe in banded iron 0-4560 Konhauser et al. (2009)
formation
A75 Europium anomaly in 0-4560 Huston and Logan (2004)
iron formation
A76 Iron isotopes in BIF and 0-4560 Johnson et al. (2008)
pyrite
A77 Iron isotopes in shale 0-4560 Johnson et al. (2008)
A78 Atmospheric oxygen 0-4560 Retallack (2007)
A79 Oxygen content of the 0-4560 Huston et al. (2004)
shallow oceans
A80 Oxygen content of the 0-4560 Huston et al. (2004)
deep oceans
A81 Number of mineral 0-4560 Hazen et al. (2008)
species
A82 Large Earth impacts 0-4560 Abbott and Isley (2002)
A83 Terrestrial impact flux 0-4560 Hartmann et al. (2000)
A84 Lunar impacts from 0-4560 Muller (2002)
spherule ages
A85 Meteor cosmic exposure 0-4560 Shaviv (2003)
ages
A86 Solar luminosity 0-4560 Gough (1981)