Plate Tectonics and Metallogeny

Laurence Robb, Department of Earth Sciences, University of Oxford, Oxford, United Kingdom
Chris Hawkesworth, School of Earth Sciences, University of Bristol, Bristol, United Kingdom
© 2020 Elsevier Inc. All rights reserved.

Introduction 1
Patterns in the Distribution of Mineral Deposits 1
Continental Growth and the Supercontinent Cycle 3
Patterns of Continental Growth 3
Supercontinent Cycles 7
Kenor 7
Nuna (also referred to as Columbia) 7
Rodinia 7
Pangea 8
Metallogeny and the Evolution of Geological Processes 8
Evolution of the Hydrosphere and Atmosphere 8
Secular Decrease in Global Heat Production and Mantle Temperature 8
Long Term Global Tectonic Trends and Mantle Convection 10
Eustatic Sea-Level Changes and “Continental Freeboard” 11
Metallogeny Through Time 11
The Archean Eon 11
The Hadean (>4000 Ma) and Eoarchean (4000–3600 Ma) stages 11
The Paleo-, Meso-, and Neoarchean stages (3600–2500 Ma) 12
The Proterozoic Eon 13
The Paleoproterozoic Era (2500–1600 Ma) 13
The Mesoproterozoic Era (1600–1000 Ma) 13
The Neoproterozoic Era (1000–541 Ma) 14
The Phanerozoic Eon 14
Plate Tectonic Settings and Ore Deposits—A Summary 16
Extensional Settings 16
Compressional Settings 17
Acknowledgments 19
References 19
Further Reading 20

Introduction

This article summarizes recent thinking that relates the formation of ore deposits to global tectonics and crustal evolution. This is
done in two ways:
1. By outlining the processes of crustal evolution and describing ore deposits in their secular and tectonic settings (such as in Meyer,
1981, 1988; Veizer et al., 1989; Barley and Groves, 1992; Windley, 1995; Groves and Bierlein, 2007; Bradley, 2011; Cawood and
Hawkesworth, 2015).
2. By describing ore deposits in the context of their host rocks and the tectonic environments in which they formed (as in Mitchell
and Garson, 1981; Hutchison, 1983; Sawkins, 1990).
Both approaches are adopted in the following sections.

Patterns in the Distribution of Mineral Deposits

Mineral deposits are not randomly distributed, either in time or in space, and there are broad patterns relating deposit types to
crustal evolution and tectonic setting. Fig. 1, for example, distinguishes between the broad-scale secular distribution of metal
deposits formed in orogenic settings and those formed in anorogenic environments and in continental basins. Deposit types classified
as orogenic (including lode-gold or orogenic gold ores, volcanogenic massive sulfides (VMS), and the porphyry-epithermal family
of base and precious metal deposits) were preferentially formed, or better preserved, in the late stages of the Archean Eon (between
3000 and 2500 Ma) and in the Phanerozoic Eon (between 541 Ma and the present day). The Proterozoic Eon, between 2500 and
541 Ma, preserves fewer of these types of deposits. By contrast, the majority of metal deposits that are associated with anorogenic

Encyclopedia of Geology, 2nd edition https://doi.org/10.1016/B978-0-12-409548-9.12535-7 1

2 Plate Tectonics and Metallogeny

Fig. 1 Distribution of ore deposit types, classified as either orogenic (upper set of histograms) or anorogenic/continental sediment-hosted occurrences (lower set
of histograms), as a function of time. The length of each histogram bar is an estimate of the proportion of ore formed over a 50 million year interval relative to the
global resource for that deposit type in the same interval. The diagram is originally after Meyer (1988) and modified after Barley and Groves (1992).
 Plate Tectonics and Metallogeny 3

magmatism (such as anorthosite hosted Ti deposits and the Olympic Dam and Kiruna type Fe oxide-Cu-Au, or IOCG, ores), as well
as sediment hosted ores (such as the SEDEX type Pb-Zn deposits and the stratiform “red-bed” type Cu deposits), appear to be
preferentially preserved in Proterozoic rocks, in particular between 2000 and 1000 Ma (Fig. 1). It seems likely, therefore, that the
nature of ore forming processes over time, as recorded in the ores that are preserved at the present day, is related to the pattern of
crustal evolution, which is, in turn, closely linked to the concept of the “supercontinent cycle.”
The episodic amalgamation and dispersal of the major continental fragments on a global scale is referred to as the supercon-
tinent cycle and exerts a fundamental control on lithospheric processes (Rogers and Santosh, 2004; Bradley, 2011; Nance et al.,
2014; Cawood and Hawkesworth, 2015). Fig. 2A illustrates the existence of 4 supercontinents over geologic time (as described in
more detail below) and the timing of their formation and dispersal compared with the approximate ages of major mineral deposits.
Porphyry and epithermal ore forming systems, for example, are not typically associated with periods of supercontinent stasis, but
rather they formed during times of collisional orogeny—specifically subduction—related with the consumption of oceanic crust as
the supercontinents coalesced. Likewise, VMS deposits tend to form in oceanic environments in which high heat flow and extension
promoted circulation of sea-water through the crust, typically during periods of extensional orogenesis. Although VMS deposits
formed throughout Earth history, it is evident that the majority occur in three intervals of time, namely in the Neoarchean at around
2700 Ma, in the Paleoproterozoic at around 1800 Ma, and in the Silurian-Devonian at around 400 Ma (Fig. 2B). In each case these
intervals coincide with orogenesis linked to assembly of three of the four supercontinents, respectively Kenor, Nuna and Pangea. The
available data suggest that no such grouping of VMS deposits accompanies the coalescence of Rodinia, which suggests that Nuna
and Rodinia may not have been significantly different from one another, and that the Mesoproterozoic Era (1600–1000 Ma) was
one when only minimal supercontinent disassembly and orogenesis occurred (Fig. 2B). Orogenic gold ores represent a diverse
group of mineral deposits that also formed throughout Earth history (Fig. 2A), but more typically in periods when supercontinents
were either assembling or breaking apart. There is a dearth of orogenic gold deposits in the Mesoproterozoic Era. Mississippi Valley
Type (MVT) Pb-Zn deposits are typically associated with convergent processes (Fig. 2A) and mineralized districts tend to form close
to craton margins and paleo-sutures where large-scale basinal fluid movement was promoted by compressive stress regimes.
In contrast to ores associated with orogenic episodes, Fig. 2A also suggests that anorogenic and continental sediment-hosted
metal deposits tend to coincide with periods of extended crustal stability and supercontinental stasis. The occurrence of
sedimentary-exhalative (SedEx) Pb-Zn-Ag deposits, for example, is largely antithetic to that of VMS ores in that many of the larger
deposits formed in the Mesoproterozoic and during the existence of Nuna. The emplacement of magmas unrelated to subduction,
such as layered mafic intrusions, kimberlites and anorogenic (or A-type) granites, all in continental settings, may be related to
superplume activity (Fig. 2A). Superplumes have been active throughout Earth history but they appear to have been more intense,
with more voluminous magma production, in Precambrian compared to Phanerozoic times (Abbott and Isley, 2002). The super-
plume era between 2500 and 2400 Ma, for example, can be linked to emplacement of intrusions such as the PGE-Cr bearing Great
Dyke as well as some of the earlier granites that host the Salobo-Carajas IOCG province. Superplume events at ca. 2050 and
1550 Ma appear to have been implicated, respectively, in the formation of the Bushveld Complex and the Olympic Dam IOCG
deposit. Finally, the long-lived Cretaceous superplume era coincided with the emplacement of kimberlites in southern Africa, many
of which are diamondiferous.
These discussions are sensitive to the nature of the geological record. The volumes of crust preserved from the Archaean (<10%
of the present volume of the continental crust) is much less than that preserved from the Phanerozoic, for example, suggesting that
large volumes of Archaean crust have been destroyed (Dhuime et al. 2012). This raises questions regarding the extent to which the
preserved geological record is therefore representative of the geological processes taking place in different stages of Earth history.
Despite the fact that there are variations in the degree of preservation (Fig. 3A), some aspects of global tectonics, such as the
formation and destruction of supercontinents, can be established despite concerns over potential biases in the geological record.
Other aspects, such as the volumes of magmas generated in different tectonic settings and how those changed over time periods
similar to the periodicity of the supercontinent cycle, may be less well preserved. Nonetheless, the secular distribution of ore
deposits of different ages in Figs. 1 and 2 is a record of what is preserved, and so the settings in which different ore deposits were
generated can be discussed without needing to infer that these settings were necessarily representative of global tectonics at the time.
The existence of numerous arc-related orogenic deposits in the late Archean, for example, reflects the preservation of greenstone
belts in stable shield areas of the world. In Proterozoic times, by contrast, longer-lived continental stability and buoyant crust might
have uplifted and eroded similar mineralized arcs so that many of the related near-surface ores were destroyed. The products of
Phanerozoic orogenies are also better preserved because many of them have not yet suffered collision, uplift and erosion and,
therefore, remain reasonably intact.

Continental Growth and the Supercontinent Cycle

Patterns of Continental Growth
The continental crust is made up of rocks of differing compositions and crust-formation ages that have survived to the present day.
Thus Archaean rocks make up <10% of the present crustal volume, although this is unlikely to be a true reflection of the volume of
continental crust at that time. Understanding the pattern of crustal evolution and metallogenesis may be informed by a knowledge
of the volumes of crust preserved from different stages in Earth history because orogenesis plays a fundamental role in the rates and
styles of ore deposit formation. Models of crustal evolution need to be devised independently of the present day distribution of
4 Plate Tectonics and Metallogeny

(A) Time (Ga)

2.9 2.5 2.0 1.5 1.0 0.5

OROGENY STASIS DISPERSAL OROGENY STASIS DISPERSAL OROGENY STASIS DISPERSAL OROGENY STASIS DISPERSAL

OROGENIC GOLD

PORPHYRY
(Cu-Mo)

EPITHERMAL
(Au-Ag)

MVT (Pb-Zn)

LAYERED MAFIC
INTRUSIONS
(PGE-Ni-Cu)

VMS (Cu-Zn)

SEDEX (Pb-Zn-Ag)

IOCG (Cu-Au)

SUPERPLUME
EVENTS

KIMBERLITE (diamond)

KENOR NUNA RODINIA PANGEA

Fig. 2 (A) Temporal distribution of the major ore deposit types as a function of the existence of supercontinents Four supercontinents are recognized, namely, from
youngest to oldest, Pangea, Rodinia, Nuna and Kenor, with decreasing levels of confidence further back in time. Mineral occurrences are illustrated as broad
histogram peaks, the relative heights of which relate approximately to frequency of occurrence and size size—the data are modified after Groves and Bierlein (2007)
and Cawood and Hawkesworth (2015). (B) Histogram illustrating the age distribution pattern of volcanogenic massive sulfide (VMS) deposits in relation to the
supercontinent cycle, and the age data are from Mosier et al. (2009).
Continued
 Plate Tectonics and Metallogeny 5

(B)
4000

VMS DEPOSITS by AGE

3500

3000

2700 Ma
2500
KENOR
Ma
2000
1800 Ma
NUNA
1500

1000
RODINIA

500
400 Ma
PANGEA

0
1 50 99 148 197 246 295 344 393 442 491 540 589 638 687 736 785 834 883 932 981 1030 1079
No. of VMS deposits

Fig. 2, Cont’d

rocks of different ages, and have been based on parameters such as the radiogenic isotope ratios of Hf in zircon and Nd in sediments.
Isotope analyses of detrital and magmatic zircons provide an accurate timepiece for the dating of many geological events, and there
are now large numbers of such analyses available worldwide. Current models for the volumes of continental crust in different eras
tend to work back in time from the present day, evaluating the proportions of new and reworked crust in intervals of time until the
point at which the first continental crust is thought to have been generated.
Curves for the increase in the volume of continental crust through time are illustrated in Fig. 3A. Early models (such as curves
1981 and 1982) advocated contrasting scenarios in which crustal growth was either very rapid early on and then slowed, or took
place in spurts at various periods, notably the late Archaean. More recent models (such as curve 2012), that are based on isotopic
data reflecting the relative proportions of juvenile versus recycled crust, suggest a more rapid pace of growth in the Archaean up to
3.0 Ga, with a lower but more constant rate thereafter (McCulloch and Bennett, 1994; Dhuime et al., 2012). A number of models
now concur that at least 70% of the present volume of the continental crust had been generated by 3 Ga, and it follows that mid- to
late-Archean times mark a reduction in the rates of crustal growth. It has been argued that this is unlikely to reflect a drop in the
actual rates at which continental crust was generated, but more likely it reflects an increase in the rates at which continental crust was
destroyed. This is, therefore, taken to be an indication of when plate tectonics became the dominant tectonic regime on Earth
(Hawkesworth et al., 2010, 2016).
The other aspect of the large number of zircon age dates now available is that detrital zircon ages define a series of well-defined
peaks and troughs (Fig. 3B). Each zircon age marks the crystallization of a magma (typically granitic), and Hf isotope ratios in
zircons indicate that the peaks of ages are associated with increased reworking of pre-existing crust (Dhuime et al. 2012). This is
attributed to crustal thickening associated with periods of continental coalescence - moreover, the peaks of ages reflect periods when
the continental crust was better preserved, not necessarily because more crust was generated but because less of the crust that had
formed was recycled or destroyed (Hawkesworth et al. 2016). The frequency distribution of detrital zircon ages therefore records the
preservation - in the sedimentary rock record—of minerals that grew in magmas associated with the amalgamation and dispersal of
supercontinents over time. The detrital zircon age data reflect cycles of formation and preservation of continental crust, but they can
also be used to assess the secular patterns of crustal growth (Fig. 3A: Dhuime et al., 2012).
It is apparent in Fig. 3B that the detrital zircon age distributions define a series of peaks that coincide broadly with the ages of
supercontinents, the latter representing the periodic assembly of most of the Earth’s continental fragments into a single, essentially
6 Plate Tectonics and Metallogeny

(A)

1.2
1978

1981

Fraction of continental crust

1.0

0.8 201
0

0.6

20
20
PHANEROZOIC

03
12
0.4

198
PROTEROZOIC

2
0.2

ARCHEAN
0.0
0.0 1.0 2.0 3.0 4.0
Ga
(B)

Nuna
Pangea Rodinia Kenor
4000

3000

2000

1500
Number

(GONDWANA)

1000
ZIRCON AGES

500

0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Ga
Fig. 3 (A) Models for crustal growth rates showing the change in ideas over time (after Korenaga, 2013), where the numbered curves indicate the year the model
was proposed. One group of models (1978, 1981 and 1982—after Fyfe, 1978; Armstrong, 1981; McLennan and Taylor, 1982) was largely based on the present day
distribution of rocks of different geological and model ages; later models (2003, 2010 and 2012—after Campbell, 2003; Belousova et al., 2010; Dhuime et al., 2012)
were based on isotopic constraints. The curve marked 2012 proposes a decrease in crustal growth rate due to increased crustal destruction rates from around
3.0 Ga and is taken to indicate the onset of conventional plate tectonics at that time. (B) Histogram showing peaks of detrital zircon ages over time (after
Hawkesworth et al., 2016) and the coincidence of these peaks with the tenure of supercontinents.

contiguous, landmass. At least 4 cycles of supercontinent formation have been recognized over the past 3 billion years (Fig. 4),
excluding Gondwana which is now considered as part of the eventual amalgamation that resulted in Pangea, rather than as a
separate supercontinent. The supercontinents depicted here are - from youngest to oldest—Pangea (which existed ca.
0.32–0.19 Ga), Rodinia (ca. 1.0–0.7 Ga), Nuna (also called Columbia, at ca. 1.8–1.4 Ga) and Kenor (ca. 2.6–2.3 Ga). If the zircon
age peaks in Fig. 3B reflect periods of enhanced preservation of crust relative to a global crustal growth rate that increased gradually
since 3 Ga, it follows that preservation of crust was best achieved during periods of lithospheric amalgamation and stability
(Hawkesworth et al., 2010).
 Plate Tectonics and Metallogeny 7

Ga
3.0 2.5 2.0 1.5 1.0 (scale not linear) 0.5 0.25 0.065 0.0

E. GONDWANA
VAALBARA
W. GONDWANA AFRICA
SUPERIA GONDWANA

KENOR- NUNA / PANGEA SOUTH AMERICA

SUPERCONTINENTS RODINIA
LAND COLUMBIA LAURENTIA
NORTH AMERICA
SCLAVIA
EURASIA
(PANNOTIA)
AUSTRALASIA

ARCHEAN PROTEROZOIC PALEOZOIC MESOZOIC CENOZOIC

PALEO- MESO- NEO-

Fig. 4 Schematic diagram showing the timing of amalgamation and dispersal of supercontinents. A number of different names have been allocated to the various
supercontinent cycles and the ones adopted here are from Hawkesworth et al. (2016).

Supercontinent Cycles
The existence of supercontinents and the cycles within which continental fragments coalesce and disperse is fundamental to all
Earth processes, including metallogeny. Early attempts to discern these cycles (Windley, 1995; Rogers, 1996) are continuously being
improved as more precise paleomagnetic and geochronological studies provide better constraints on apparent polar wander paths.
The pattern of crustal evolution in the Phanerozoic Eon is well understood, involving the progressive amalgamation of large
continental fragments to form the supercontinent of Pangea. The latter existed during Permian and Triassic times and was
subsequently dispersed to form the present day continental geography (Fig. 4). A relatively high degree of confidence marks the
reconstruction of continental configuration during the Phanerozoic Eon and there is general agreement on the paleogeography of
land-masses such as Gondwana, Laurentia, and Pangea. As one goes back in time into the Precambrian, however, the situation
becomes progressively more uncertain and disagreement often accompanies the reconstruction of supercontinent geometry.
Despite these difficulties, progress has been made in reconstructing continental paleogeography back through geological time
although, as mentioned, the nature of supercontinent cyclicity in the Proterozoic and Archean Eons remains tenuous. The historical
development of the concept of supercontinent cyclicity is described by Nance and Murphy (2013) and Fig. 4 presents a scheme
suggesting the existence of 4 supercontinents since approximately 3.0 Ga. The following summary is based on this fourfold
subdivision, from oldest to youngest.

Kenor
The oldest amalgamated, but now dispersed, continental land-mass was suggested to have comprised the ancient Kaapvaal and
Pilbara cratons of southern Africa and Western Australia, as well as parts of India and Antarctica, and it is now generally referred to as
Vaalbara (Button, 1979; Cheney, 1996). Vaalbara is believed to have coalesced from about 3000 Ma but had broken up well before
2100 Ma and prior to emplacement of the Bushveld Complex on the Kaapvaal Craton. Other continental amalgamations, in the
form of Superia and Sclavia (Fig. 4), are also suggested to have existed in Neoarchean and Paleoproterozoic times (Bleeker, 2003;
Nance et al., 2014) and it is the amalgamation of all these fragments that may have resulted in the existence of an early
supercontinent called Kenor (or Kenorland) between around 2.6 and 2.3 Ga (Williams et al., 1991).

Nuna (also referred to as Columbia)

The Mesoproterozoic Era is believed to have included a more robust supercontinent referred to as either Nuna (an acronym for
Northern Europe and North America) or Columbia (Piper, 1976; Hoffman, 1988; Park, 1995). There are approximately 35 known
fragments of Archean-aged continents preserved on Earth at present (Bleeker, 2003) and these would appear to have been largely
cohesive at around 1.8–1.7 Ga, to form what some regard as the first genuine supercontinent, Nuna. Nuna is believed to have been
in existence for around 400 million years (between ca. 1.8 and 1.4 Ga), through late Paleoproterozoic into Mesoproterozoic times
(Zhao et al., 2004; Zhang et al., 2012; Evans and Mitchell, 2011).

Rodinia
There are indications that a Neoproterozoic supercontinent, now referred to as Rodinia (McMenamin and McMenamin, 1990),
started to assemble from about 1200 Ma, and then dispersed again by about 700 Ma. Rodinia, like Nuna, was a long-lived entity
providing lithospheric stability between approximately 1.0 and 0.7 Ga. The supercontinent broke up episodically over a protracted
period that may have exceeded 200 million years. The break-up and dispersal of Rodinia coincided with the advent of catastrophic
climate change and dramatic biological diversification associated with the so-called “Snowball Earth” events of the Cryogenian and
Ediacaran Periods of the Neoproterozoic Era.
8 Plate Tectonics and Metallogeny

Pangea
The most recent supercontinent, Pangea, coalesced from the fragments of Rodinia, and assembled as Laurasia (a combination of
Laurentia and Eurasia) and Gondwana re-united by progressive subduction of the Rheic Ocean in late-Paleozoic and Mesozoic
times. The geological, paleontological and paleomagnetic evidence for the existence of a combined landmass in Permian-Triassic
times is robust, but the details of how it was assembled are convoluted (Torsvik and Cocks, 2013). In late Devonian times, at
370 Ma, the large continental landmass of Gondwana was centered on the South Pole and separated from Laurasia by the Rheic
Ocean (Fig. 5A). Collision of Laurasia and Gondwana occurred in late Carboniferous times (ca. 320 Ma; Fig. 4) to form Pangea, a
supercontinent that initially existed largely in the southern hemisphere. Pangea drifted northwards as a cohesive block such that in
the early Mesozoic, at 250 Ma (Fig. 5B), significant portions of the supercontinent lay at equatorial latitudes. This explains why
significant accumulation of lignite and coal developed at that time in sedimentary sequences amenable to peat accumulation. It also
explains the formation of extensive evaporitic sequences and petroleum source rocks in marginal marine settings associated with the
Tethys Ocean at, and subsequent to, this era (Scotese, 2009; Torsvik and Cocks, 2013). These observations attest to the relationship
between plate motions and the creation of environmentally sensitive ore forming niches.
Pangea was not a long-lived entity and it had already started to break-up in the early Jurassic Period. Major fragmentation
commenced when Laurasia separated from West Gondwana at ca. 195 Ma, to form the incipient Atlantic Ocean. Further break-up
separated West Gondwana from East Gondwana to form the Indian Ocean from 170 Ma onwards. Pangea has now dispersed and its
remnants occur as the five major landmasses that form the present day continental geography (Fig. 4). Some continents are currently
undergoing reassembly, with collision of India with Asia and the partial amalgamation of Africa and Europe, having been evident
over the past 50 million years.

Metallogeny and the Evolution of Geological Processes

There are a number of fundamental geological processes that influence the pattern of global metallogeny. Some change progres-
sively with time, such as the evolution of the hydrosphere and atmosphere, the secular decrease in global heat production, and
perhaps the nature of mantle convection. Others are more episodic, including eustatic sea level changes, creation of oceanic crust
and, as discussed in the previous section, long-term global tectonic trends reflected in the generation and fragmentation of
supercontinents. These are briefly discussed below, with an explanation of how and why they influence the formation of ore
deposits.

Evolution of the Hydrosphere and Atmosphere

The changing budget of O2, and to a lesser degree CO2, in the Earth’s atmosphere with time has played a role in the formation of ore
deposits, especially those related to redox processes and to the weathering and erosion of continental crust. In Archean times the
atmosphere contained very little free molecular oxygen. A reduced atmosphere in the Archean helps to explain many of the features
of ore formation at that time, including the widespread mobility of Fe2+ and development of Algoma and Superior type banded
iron-formations, as well as the preservation of detrital grains of uraninite and pyrite in sedimentary sequences of the Witwatersrand
and Huronian basins. The transition from the Archean to the Proterozoic Eon, at 2500 Ma, broadly coincides with the beginnings of
an increase in atmospheric oxygen (Fig. 6), with the Great Oxidation Event (GOE) having now been dated at ca. 2.3 Ga (Bekker
et al., 2004). This event is possibly related to the evolution of primitive life and the widespread proliferation of cyanobacteria
capable of photosynthetically producing oxygen.
The interval between 2500 and 2000 Ma is known as the period of “oxyatmoinversion” and is believed to coincide with a
significant rise in atmospheric oxygen levels (Holland, 2006). It is the period after which development of Algoma and Superior type
banded iron-formations (BIFs) and related bedded manganese deposits generally ceased. Another significant increase in atmo-
spheric oxygen levels coincided with the end of the Proterozoic Eon, when macroscopic, multicellular life forms (Metazoan fauna
which require oxygenated respiration) proliferated. In addition, the early Phanerozoic is also the time when vascular plants evolved
and this, too, would have contributed to the increase in atmospheric oxygen levels (Fig. 6). In contrast to the increasing levels of O2
in the atmosphere, CO2 has progressively decreased with time (Mojzsis, 2001). The high levels of CO2 in the early atmosphere were
a product of extensive volcanism and outgassing during the Archean, but levels have dropped steadily, but episodically, as a result of
carbonate deposition in sediments. The evolution of the Earth’s atmosphere and, in particular, the global O2 and CO2 budgets, are
essentially non-recurrent and irreversible and contribute to the time-specific occurrence of ore deposit types such as banded iron-
formations and Witwatersrand type placers (Fig. 6).

Secular Decrease in Global Heat Production and Mantle Temperature

The amount of heat produced from the decay of long-lived isotopes such as U, Th, and K, in the Earth was significantly greater (by a
factor of two to three times) in the Archean than it is today. This pattern, like that for atmospheric evolution, is also non-recurrent
and irreversible. There are several consequences for magma generation of the secular decrease in heat production. One of the most
obvious is that the Archean mantle was hotter (by about 250  C; Windley, 1995; Herzberg et al., 2010) than at present and this is
likely to have resulted in higher volumes and degrees of melting at that time. A hotter mantle accounts for the production of the
high-Mg komatiites that typify many Archean greenstone belts and which are relatively rare in younger oceanic settings. Higher heat
 Plate Tectonics and Metallogeny 9

(A)

370 Ma LATE DEVONIAN

Land
South
Shallow Shelf China
Deep Shelf
Ocean
Evaporites
Pa
lae
Arabia ote
thy
s

GONDWANA

India

r
South

asca
Pole

ag
Mad
South
America Australia
South Africa

East
Antarctica

Rh
eic
Oc 60°S
ea
n

r
Laurasia

to
ua
°S
30

Eq
(B)

North
China
Eq
ua
to
r

30°S
Laurentia
tor

NW
ua

NE
Eq

Africa
Africa

India
PA N G E A

South
America West
South Australia
Africa
East
Antarctica

South
280 Ma Pole
EARLY PERMIAN
Land
Shallow Shelf
Deep Shelf NZ
Ocean
MBL
Evaporites
TI
Ice cap Patagonia °S °S
60 30

Fig. 5 (A) Continental paleogeography in the late Devonian Period, at 370 Ma. Gondwana was a very large continental landmass, but not a supercontinent,
centered on the South Pole; (B) reconstruction of Pangea at 280 Ma (after Torsvik and Cocks, 2013).
10 Plate Tectonics and Metallogeny

NEO-PRO-
TEROZOIC
MINIMAL FREE PHANEROZOIC –
SNOWBALL
OXYGEN PHOTOSYNTHESIS – EARTH – PROLIFERATION
LITTLE OR NO FREE OXYGEN CONSUMED BY FREE OXYGEN INTO RISE IN OF PLANTS,
OXIDATION OF ATMOSPHERE ATMOSPHE- MULTICELLULAR
IRON RIC ORGANISMS
OXYGEN

ALGOMA BIF SSC

(Cu-Co-Ag)
SUPERIOR
BIF
SUPERIOR POI
QUARTZ PEBBLE GIF
CONGLOMERATE RAPITAN
(pyrite-gold-uraninite)
(pyrite- gold- uraninite) BIF

0.3
Atmospheric PO2 (atm)

0.2
GREAT OXIDATION EVENT

EN
YG
0.1

OX
ATMOSPHERIC

0
3.8 3 2 1 0
Ga

Fig. 6 Estimated rise of atmospheric oxygen over the past 3.8 Ga on the basis of 5 evolutionary stages, after Holland (2006). Also shown are the approximate age
ranges over which ore deposit types sensitive to atmospheric oxygen levels and redox processes formed.

production thus provides an explanation for the preferential occurrence, in the Archean, of komatiite-related magmatic Ni-Cu
sulfide deposits, such as those at Kambalda, Western Australia and in Zimbabwe. Another consequence for higher heat flow in the
Archean is that convective overturn in the mantle may have been more rapid (Cawood et al., 2018) which has implications for the
mechanisms of crustal growth (as depicted, for example, in Fig. 7) and the types of mineral deposits forming at that time.

Long Term Global Tectonic Trends and Mantle Convection

Unlike the patterns imposed by atmospheric evolution and heat production, the supercontinent cycle is recurrent and cyclical in
that continental fragments have amalgamated and dispersed a number of times through the history of the Earth (Fig. 4). This
cyclicity helps explain why some ore deposit types, or at least the processes that give rise to them, are recognizably repetitive despite
the preferential preservation of ore deposits in younger rocks. Fig. 7 illustrates one model showing how the lithosphere may have
changed with time. The diagram depicts a pre-3200 Ma “stagnant lid” regime in which relatively thin, granite-greenstone crust (i.e.,
comprising continental fragments of TTG (tonalite-trondhjemite-granodiorite) type crust and mafic material represented by
greenstone belts) is underlain by hot mantle undergoing rapid, small-scale convection. Formation of magma and new crust during
stagnant lid processes occurred not so much by “subduction” in the sense of conventional plate tectonics, but by processes that vary
from remelting the base of a magmatically-thickened volcanic crust, to others that include foundering of more dense volcanic
assemblages—sometimes referred to as “drip” tectonics. This regime envisages, simplistically, that a basalt-dominated stagnant lid is
periodically over-turned back into the mantle with accompanying partial melt events. The change to a more rigid, and yet more
mobile lithosphere is believed to have been transitional and occurred between ca. 3200 and 2500 Ma (Fig. 7; Bédard, 2018; Cawood
et al., 2018). This transitional period could have marked the beginnings of sustainable plate tectonics. Subduction thickens the
continental crust, both magmatically and tectonically, with lithosphere stabilization often accompanied by the onset of potassic
granite magmatism. The development of more rigid continental lithosphere, and its differentiation from oceanic domains, results in
the onset of sustainable plate tectonics in the period 3200–2500 Ma (Fig. 7) and is supported by features such as (i) the appearance
of eclogitic inclusions in diamond at around 3.0 Ga (Shirey and Richardson, 2011); (ii) the development of major dyke swarms;
(iii) the appearance of potassic granites; and (iv) intracratonic sedimentation (such as the Witwatersrand Basin) in many parts of the
world (Cawood et al. 2018).
 Plate Tectonics and Metallogeny 11

> 3200 Ma 3200-2500 Ma <2500 Ma

‘stagnant lid’ processes accretionary orogens collisional orogens and supercontinents
TTG early mafic crust calc-alkaline andesitic crust

cratonization CONTINENTAL
ics
LITHOSPHERE cton
t e OCEANIC
te
pla LITHOSPHERE

small wavelength ‘drip’

mantle convection tectonics

large wavelength
mantle convection

decreasing mantle temperature

Fig. 7 Schematic illustration depicting the evolution of mantle convection, tectonic style and crustal thickness over geological time (modified after Cawood et al.,
2018).

Eustatic Sea-Level Changes and “Continental Freeboard”

Continental freeboard, or the relative elevation of the continental land masses with respect to sea level or the geoid, is effectively a
measure of the area of the exposed continents compared to that covered by the oceans. Continental freeboard is reduced when the
average ocean depth becomes shallower, since this causes a marine transgression and flooding of continental margins. Such
transgressions are linked to events of large-scale continental dispersal, which in turn reflect enhanced creation of oceanic crust,
and active magmatism and uplift along mid-ocean ridges. By contrast, an increase in the area of the continents occurs when the
oceans are deepened and marine regression exposes the continental shelf—this situation tends to be associated with periods of
tectonic stability and muted mid-ocean ridge magmatic activity (Worsley et al., 1984; Nance et al., 1986). The cyclicity of water
depth at the edge of continental shelves, and by implication continental freeboard, in the Phanerozoic Eon is well documented
(Fig. 8) and shows that a period of maximum continental exposure and ocean low-stand (i.e., deep oceans) coincides with the
existence of Pangea in Permian-Triassic times. The nature and duration of continental dispersion and amalgamation in the
Phanerozoic Eon, and their projection into the future, are also shown in Fig. 8. A similar period of ocean lowstand may also
have occurred at the time of an earlier supercontinent, Rodinia, at approximately 1000 Ma. From a metallogenic viewpoint,
continental freeboard has its biggest influence on the nature and preservation of sediments forming on the continental shelves.
Marine transgressions flood the continental shelf, preserving the sediments within which deposits such as heavy mineral beach
placers, SEDEX Pb-Zn ores, banded iron-formations, bedded manganese deposits, and phosphorites might have formed. Oceanic
lowstand and exposure of continental margins, on the other hand, result in shelf erosion and possible destruction of any ore
deposits present in the associated sedimentary sequences.

Metallogeny Through Time

The following sections summarize the nature of mineral deposits in terms of Earth’s geodynamic evolution, subdivided and
described on the basis of the three major geological time periods (or Eons), namely Archean, Proterozoic and Phanerozoic.

The Archean Eon

Given the relatively poor degree of preservation and exposure of Archean continental crust it is difficult to reconstruct the
paleogeography of ancient crustal fragments with any certainty, but the tectonic processes that formed Archean crust are relatively
well constrained. Archean metallogeny is, therefore, described in terms of the tectonic processes applicable at the time and is
classified in terms of its major chronostratigraphic subdivisions, each of which has a specific metallogenic character.

The Hadean (> 4000 Ma) and Eoarchean (4000–3600 Ma) stages
The Hadean Era refers to that period of Earth history for which there is very little evidence in the rock record and which is nominally
pre-Archean (i.e., pre-4000 Ma). It was a time of global differentiation and accretion, as well as intense meteorite bombardment.
There is also evidence to suggest that the early atmosphere and the precursors to the present oceans formed only at the end of the
Hadean era, once the main period of accretion and meteorite bombardment had terminated at ca. 3900 Ma (Kasting, 1993).
The implications for metallogenesis are that sedimentary and hydrothermal process are likely to have been inconsequential in the
12 Plate Tectonics and Metallogeny

Fig. 8 Pattern of supercontinent cycles, reflecting assembly and fragmentation of Pangea during the Phanerozoic Eon (and into the future), and showing water
depth of the continental shelf (after Nance et al., 1986).

Hadean, and any ore deposits that did form at that time were, therefore, largely igneous in character. It is conceivable, for example,
that oxide and sulfide mineral segregations accumulated from anorthositic and basaltic magmas at this time. However, the only
preserved record of such rocks within reach of humankind at present is on the Moon.
The Eoarchean refers to the dawn of Archean time and to rocks formed prior to 3600 Ma, but subsequent to 4000 Ma. The best
preserved section of Archean crust that falls into this time bracket is the 3800 Ma Isua supracrustal belt and associated Itsaq
(previously called Amitsoq) gneisses of western Greenland. The Isua belt comprises mafic and felsic metavolcanics, as well as
metasediments, and resembles younger greenstone belts from elsewhere in the world. Although only 4  30 km in dimension, the
Isua belt contains a major chert-magnetite banded iron-formation component as well as minor occurrences of copper-iron sulfides
in banded amphibolites and in iron-formations (Appel, 1983). The largest iron-formation contains an estimated 2 billion tons of
ore at a grade of 32% Fe. Scheelite mineralization has also been found in both amphibolite and calc-silicate rocks of the Isua belt, an
association which suggests a submarine-exhalative origin. The coexistence of banded iron-formations and incipient volcanogenic or
sedimentary exhalative, massive sulfide deposits points to an oceanic environment and, in the latter case, processes involving
circulation of seawater through oceanic crust. Although the zones of known mineralization in the Isua belt are sub-economic, at
3800 years old they represent the oldest known ore deposits on Earth.

The Paleo-, Meso-, and Neoarchean stages (3600–2500 Ma)

These stages of Archean crustal evolution took place over an extended duration of more than 1000 million years, during which time
there appears to have been a transition from a stagnant lid and drip tectonic regime to one in which a more conventional style of
plate tectonics might have developed (Fig. 7). Evidence exists through the Meso- and Neoarchean Eras for substantial crustal
amalgamations, such an early Vaalbara continent and later Superia and Sclavia continents (Bleeker, 2003).
The existence of Vaalbara receives support from the similarities that exist in the nature and ages of Archean greenstone belts and
supracrustal sequences on the Pilbara and Kaapvaal cratons (Cheney, 1996; Martin et al., 1998), a feature that is especially striking
when comparing the Superior-type banded iron-formations of the two regions. It was previously thought that the Archean
Witwatersrand basin on the Kaapvaal Craton was a unique occurrence, but exploration in Western Australia, has recently revealed
the existence of sedimentary sequences of similar age that also appear to be well endowed with gold mineralization.
The Neoarchean era represents a period of significant crustal growth and the development of abundant mineralization. The
processes active from around 2800 Ma were not unlike those taking place later in Earth history, involving plate subduction, arc
magmatism, continent collision and rifting, and cratonic sedimentation. A wide variety of ore-forming processes therefore
characterizes these periods of Earth history. Well mineralized examples of continental crust that formed in the period
2800–2500 Ma are represented by the granite-greenstone terranes of the Superior Province of Canada, as well as the Yilgarn and
Zimbabwe cratons. Greenstone belts formed from arc-related volcanism host important volcanogenic massive sulfide (VMS) Cu-Zn
ore bodies, such as those at Kidd Creek and Noranda in the Abitibi greenstone belt of the Superior Province. Off-shore, in more
 Plate Tectonics and Metallogeny 13

distal environments, chemical sedimentation gave rise to Algoma type banded iron-formations, examples of which include the
Adams and Sherman deposits, also in the Abitibi greenstone belt. Greenstone belts formed at this time also comprise komatiitic
basalts that, under conditions favorable for magma mixing and contamination, exsolved immiscible Ni-Cu-Fe sulfide fractions to
form deposits such as Kambalda in Western Australia and Trojan in Zimbabwe. During and soon after periods of compressive
deformation, major suture zones became the focus of hydrothermal fluid flow derived either from metamorphic devolatilization or
late-orogenic magmatism. This resulted in the formation of the varied styles of orogenic gold mineralization that are typical of most
Meso- and Neoarchean granite-greenstone terranes worldwide. Examples include important deposits such as the Golden Mile in the
Kalgoorlie district of Western Australia and the Hollinger-McIntyre deposits of the Abitibi greenstone belt.
Early intracratonic styles of sedimentation, often in foreland basinal settings, gave rise to concentrations of gold and uraninite
represented by the ca. 3.0–2.7 Ga Witwatersrand basin in South Africa. At least some of this mineralization is placer in origin and
was derived by eroding a fertile Archean hinterland that appears to have been elevated compared to the adjacent basin. The passive
margins to these early continents would have developed stable platformal settings onto which laterally extensive Superior type
banded iron-formations were deposited. A very significant period for deposition of iron ores such as those of the Hamersley and
Transvaal basins of Western Australia and South Africa respectively, as well as the Mesabi range of Minnesota, seems to have been
around the Archean-Proterozoic boundary at 2500 Ma, by which time continental crust was both thick and increasingly rigid.

The Proterozoic Eon

The period of time subsequent to 2500 Ma represents a major transition in the nature of crustal evolution, involving changes in the
volume and composition of the continents, tectonic regimes, and atmospheric make-up, all of which affected ore-forming processes
and characteristics (Fig. 1). The Proterozoic Eon spans a vast period of geological time, from 2500 to 541 Ma, including the period
between 1800 and 1000 Ma that was marked by a relative paucity of orogenic and/or magmatic-hydrothermal deposit types, but
abundant ores hosted in intracontinental sedimentary basins and anorogenic igneous complexes. The reasons for this pattern are
multifaceted, but, as a generalization, they are related to a higher degree of continental stability and the existence of long-lived
supercontinents such as Nuna and Rodinia (Zhang et al., 2012; Nance et al., 2014). A substantial volume of continental crust
appears to have been in existence by the beginning of the Proterozoic Eon (perhaps 70% or more of the present volume, Fig. 3A)
and it is widely held that the first true supercontinent, Nuna, came into existence during this time.

The Paleoproterozoic Era (2500–1600 Ma)

From a metallogenic perspective, the period of earth history between 2500 and 1600 Ma is significant because of the major changes
that occurred to the atmosphere, especially the rise in atmospheric oxygen levels at around 2300 Ma (Fig. 6). Prior to this time, the
major oxygen sink was the reduced deep ocean where any photosynthetically produced free oxygen was consumed by the oxidation
of volcanic gases, carbon, and ferrous iron. In this environment banded iron-formations, as well as bedded manganese ores,
developed, as indicated by the widespread preservation of both Algoma and Superior type iron deposits. The increase in ferric/
ferrous iron ratio in the surface environment that accompanied oxyatmoinversion at 2300 Ma, and the accompanying depletion in
the soluble iron content of the oceans resulted in fewer BIFs forming after this time (Bekker et al., 2010, 2014). The stability of easily
oxidizable minerals such as uraninite and pyrite is also to a certain extent dependent on atmospheric oxygen levels and it is,
therefore, relevant that major Witwatersrand-type placer deposits did not form after about 2000 Ma.
Metallogenic patterns during the Paleoproterozoic Era were dominated by wide-ranging orogenic processes accompanying plate
movements associated with the break-up of Superia and Sclavia and the assembly of Nuna between ca. 1800 and 1400 Ma. The
break-up of Superia was accompanied, between 2000 and 1700 Ma, by the creation of new oceanic crust and the formation of
volcanogenic massive sulfide Cu-Zn deposits such as Flin Flon in Canada, Jerome in Arizona, and the Skellefte (Sweden)-Lokken
(Norway) ores of Scandinavia. At 2055 Ma on the Kaapvaal craton, the enormous Bushveld complex with its world-class PGE, Cr,
and Fe-Ti-V reserves was emplaced, as was the Phalaborwa alkaline complex with its contained Cu-P-Fe-REE mineralization—both
generated in intraplate settings. In West Africa the period between 2100 and 1900 Ma saw the development of substantial juvenile
crust during the Eburnean orogeny, accompanied by the formation of extensive orogenic gold mineralization.
The amalgamation of Nuna was followed by a long period of cratonic stability that resulted in the deposition, between 1800 and
1500 Ma, of marginal marine sedimentary basins that host the important SEDEX Pb-Zn ores of eastern Australia (Mount Isa, Broken
Hill, and McArthur River) and South Africa (Aggeneys and Gamsberg). Anorogenic magmatism was widespread in Nuna times—in
South Australia, for example, the 1590 Ma Roxby Downs granite-rhyolite complex (Johnson and Cross, 1995), host to the
enormous magmatic-hydrothermal Olympic Dam iron oxide-Cu-Au-U deposit, was emplaced. Anorogenic granite magmatism at
1880 Ma may also have given rise to the later stages of IOCG style mineralization (such as Estrela; Volp, 2005) in the giant deposits
of Carajas, Brazil.

The Mesoproterozoic Era (1600–1000 Ma)

The dispersal of Nuna and re-assembly of Rodinia are events assigned to the Mesoproterozoic, and evidenced globally by the
presence of the Grenvillian orogenies between 1400 and 1000 Ma. However, the processes that created Rodinia have preserved few
world-class ore deposits—the reasons for this are enigmatic but possibly related, in part, to the longevity and stasis of both Nuna
and Rodinia and the erosive removal of buoyant mineralized upper crust. The deposits that formed and are preserved from the
Mesoproterozoic Era are often termed “anorogenic” as they are not typically related to plate movement and subduction. Large
volumes of anorogenic magmatism emplaced in the period 1500–1000 Ma are associated with a number of mineral deposits in a
14 Plate Tectonics and Metallogeny

belt stretching from southern California through Labrador into Scandinavia (Windley, 1995). For example, gabbro-anorthosite
intrusions host the large magmatic Fe-Ti (ilmenite) ore bodies of the Marcy massif in the Adirondacks and the ca. 1060 Ma Lac Tio
deposit in Quebec, as well as the 920 Ma Tellnes deposit of SW Norway (Charlier et al., 2015). The same belt also contains alkali
granite-rhyolite complexes which give rise to Fe-Au-REE resources such as those of the ca. 1480 Ma St Francois mountains of
Missouri (Denison et al., 1984). This basement was eroded to form the sediments of the 1440 Ma Belt basin in the northwest USA,
host to the Sullivan Pb-Zn SEDEX deposit in British Columbia, Canada. In addition, intracontinental rifting at around 1100 Ma in
ancestral Rodinia, gave rise to the formation of the 2000 km long Keweenawan mid-continental rift, stretching from Michigan to
Kansas and filled with a thick sequence of bimodal basalt-rhyolite volcanics overlain by rift sediments. The latter form the host rocks
to the stratiform Cu-Ag White Pine deposit in Michigan.

The Neoproterozoic Era (1000–541 Ma)

The Neoproterozoic commenced with the formation, from around 1000 Ma, of the supercontinent Rodinia. Like Nuna, Rodinia
was long-lived and only started to partially fragment after a static period of more than 250 million years. Substantial parts of what
had been Rodinia then reconvened toward the end of the Proterozoic (at 541 Ma) to form the Gondwana land-mass, as evident, for
example, in the Pan-African and Brasiliano orogenies. In detail, the assembly of Gondwana was long-lived, and the time between
750 and 550 Ma was characterized by the development of at least two, and in places possibly four, major ice ages, one or more of
which was near global in cover and extended to equatorial latitudes. The concept of the Neoproterozoic “Snowball Earth” (Harland,
1965; Hoffman et al., 1998) has important implications for metallogeny as well as the proliferation and diversification of organic
life at the Precambrian—Cambrian boundary.
The major ore deposits of the Neoproterozoic reflect the conditions of Gondwana continental stability, as well as the periods of
near global ice cover and attendant anoxia, that prevailed at this time. The extensively developed ironstone ores of northwest
Canada and South Australia, associated with the 750–725 Ma Rapitan and Sturtian glaciogenic rocks respectively, are considered to
be the result of the build-up of ferrous iron derived from off-shore hydrothermal vents in the reduced ocean waters that
accompanied the development of vast continental and oceanic ice sheets at this time. In a similar vein, the Precambrian-Cambrian
boundary at around 540 Ma is characterized by the first major global phosphogenic event that resulted in the development of
significant accumulations of phosphatic sedimentary rock (phosphorites) at equatorial paleolatitudes in places such as Mongolia
and Queensland, Australia (Cook and Shergold, 1984; Filippelli, 2008). As with sedimentary iron ores, phosphorites reflect the
upwelling of deep, nutrient-rich ocean waters onto shallow continental shelves with the syn-sedimentary precipitation of carbonate-
apatite (or collophane) onto the shelf floor.
The formation of the extensive, stratiform Cu-Co clastic sediment hosted ores of the Central African Copperbelt is also
considered to have formed in an environment influenced by the Snowball Earth. The host Katangan sediments were deposited
on a fertile Paleoproterozoic basement (Master et al., 2005) in what was initially an intracontinental rift, the development of which
overlapped with both the Sturtian and Marinoan glacial events. The Grand and Petit Conglomerates of the Katangan sequence, for
example, represent glaciogenic sediments capped by carbonates which are correlated, respectively, with the Sturtian and Marinoan
events (Windley, 1995; Master et al., 2005). The implications for mineralization are in the development of a significant volume of
reducing sediments that cap an oxidized package within which Cu and Co were soluble. Influx of Cu and Co-rich solutions, perhaps
derived from the local basement, occurred as diagenetic fluids migrated along growth faults and through the basin at times that were
essentially coeval with basin deposition. Precipitation of metals occurred when the metal-charged oxidized fluids encountered
overlying, reduced sediments or fluids, the latter representing the products of Snowball Earth-related anoxia (Greyling et al., 2005).
The proliferation and diversification of organic life in late Neoproterozoic times also resulted in the first substantial develop-
ment of highly carbonaceous sediments. A good example of these are the Ediacaran Hormuz sediments that flank the eastern
Arabian shield and which represent important source rocks for the vast Mesozoic oil and gas fields of the Arabian Gulf
(Windley, 1995).

The Phanerozoic Eon

By comparison with earlier eons, crustal evolution and tectonic settings in the Phanerozoic are well understood. A significant
proportion of the Phanerozoic Eon was characterized by geological processes that reflect a supercontinent cycle during which the
dispersal of Gondwana in the early Paleozoic was followed by the amalgamation of continental material to form the Pangea
supercontinent in early Mesozoic times. The remainder of the Phanerozoic reflects the start of another cycle involving the dispersal
of Pangea to form the present day continental geography. It seems likely that the dispersal of Pangea is now well advanced (Nance
et al., 1986) and that current plate movements will result in the reassembly of a significant proportion of the continents over the
next 100–200 million years (Fig. 8).
The assembly of Pangea took place throughout the Paleozoic Era with the culmination of continental amalgamation in Permo-
Triassic times. This was a relatively short-lived supercontinent, however, and beak-up commenced at around 230 Ma with increased
global volcanism (e.g., the Siberian continental flood basalt province and the central Atlantic magmatic province), increased
atmospheric greenhouse gas production and enhanced deposition of petroleum and coal bearing sediments (Fig. 9).
An important product of Pangea break-up in late Mesozoic and early Cenozoic times was the development of new crust that
accompanied the Cordilleran and Andean orogenies along the western margins of North and South America. This extensive
orogenic belt, extending along the entire ocean-continent interface of western Pangea, occurred in response to subduction of the
original Panthalassa Ocean (now the eastern Pacific) beneath Laurentia (now North America) and part of west Gondwana (now
 Plate Tectonics and Metallogeny 15

400

HA
LL SEA LEVEL
AM
300

Sea Level Changes (m)

Pangea
200

CU
RVE
100

SUPERPLUME

SUPERPLUME
SUPERPLUME
0

-100

PANNOTIA /GONDWANA PANGEA

Supercontinent
FRAGMENTATION DISPERSAL ASSEMBLY STASIS FRAGMENTATION
cycle

Fossil fuels
Petroleum
Coal

P
A
Volcanogenic N
massive sulfide G
Cu -Zn
E
A

Clastic hosted
(Red-bed) Cu(Ag)
(SEDEX) Pb-Zn-Ag

Marine black shale

Oolitic Fe

Superplumes
CAMB. ORD. SIL. DEV. CARB. PER. TR. JUR. CRET. CEN.
600 500 400 300 200 100 0
Ma

Fig. 9 Occurrences (in terms of relative abundances) of a variety of different ore types with time and with respect to the Phanerozoic supercontinent cycle
(modified after Titley, 1993; with information from Nance et al., 1986; Larson, 1991; Hallam, 1992; and Barley et al., 2005).
16 Plate Tectonics and Metallogeny

South America). The enormous continent-floored magmatic arc that resulted from this subduction gave rise to the 130–80 Ma
granite batholiths and felsic volcanic rocks of the Cordillera. Ongoing subduction during the Laramide orogeny (80–40 Ma)
continued to feed the magmatic arc and numerous, metallogenically important, I-type granite batholiths were emplaced at this time.
In South America, Andean evolution was analogous to that further north, but characterized by smaller degrees of accretion and
crustal extension. This arc was the site of protracted magmatism that built the Western Cordillera in North America, and in the
Andes gave rise to the volcanic edifices which host the giant porphyry and epithermal Cu-Mo-Au-Ag deposits of Peru, Bolivia,
and Chile.
The relatively well defined global tectonic cycles of the Phanerozoic Eon are also clearly reflected in secular metallogenic trends.
Titley (1993), for example, noted that the distribution of stratabound ores (i.e., volcanogenic massive sulfide (VMS), clastic
sediment hosted Pb-Zn (SEDEX), and Cu (red-bed) deposit types) could be related to the tectonic cycles of Pangea amalgamation
and break-up (Fig. 9). Preferential development of VMS deposits appears to be associated with two periods in the supercontinent
cycle of elevated sea level (highstand) associated with continental dispersal, namely, after Gondwana break-up in the early
Paleozoic and during the dispersal of Pangea in Mesozoic times. This association is considered to reflect the processes of rifting,
enhanced ocean crust production, and hydrothermal exhalation that accompany supercontinent dispersal. Similar patterns are also
evident in the paleo-equatorial accumulation of organic-rich shales and oolitic ironstone ores, which preferentially occur in the
same two intervals, namely after Gondwana break-up in the Ordovician-Devonian and post-Pangea in Jurassic-Cretaceous times
(Fig. 9). In these cases, increased exhalative activity, carbon dioxide production, global warming, and organic productivity are
interrelated processes that result in suitable conditions for black shale and ironstone precipitation in the oceans. In contrast, clastic
sediment hosted base metal ores of the SEDEX Pb-Zn-Ag and red-bed Cu types tend to form at different stages of the tectonic cycle.
These ores are preferentially developed at the time of maximum continental amalgamation and stasis, namely during the existence
of Gondwana over the Precambrian-Cambrian transition, and in Pangea times during the late Paleozoic-early Mesozoic (Fig. 9).
In terms of processes, these ore types appear to have formed preferentially at sites of continental rifting and during sea-level
lowstand (Titley, 1993).
It is pertinent to note that Phanerozoic superplume events are also related in time to periods of enhanced organic productivity
and formation of fossil fuels (Fig. 9). The superplume events are marked by enhanced volcanic activity, and appear to be associated
with periods of greenhouse gas emission and global warming, which, in turn, stimulate organic productivity. The mid-Cretaceous
superplume, active between 120 and 80 Ma, coincides with the formation of voluminous organic-rich shales deposited largely in
the low-latitude Tethyan seaway. These shales are thought to be genetically linked to some 60% of the world’s oil reserves (Larson,
1991). The late Carboniferous-Permian superplume is similarly associated with deposition of a large proportion of the world’s coal
reserves, formed between 320 and 250 Ma, again at a time when organic productivity increased and tropical, swampy conditions
prevailed along flooded continental margins associated with the corresponding sea-level rise. Mantle plumes, have also been
invoked for the preferential development of alkaline and kimberlitic magmatism associated with enhanced igneous activity in the
mid-Cretaceous and again in the Cenozoic (Chalapathi Rao and Lehmann, 2011). Many of these intrusions, important for Cu-REE-
P-Fe-Au mineralization and also diamonds, are found on old, stable cratons and located along ancient lineaments that might have
been reactivated during extension and crustal thinning associated with hotspot activity.

Plate Tectonic Settings and Ore Deposits—A Summary

The description of ore deposits in their plate tectonic settings has been carried out in considerable detail in numerous publications
(Mitchell and Garson, 1981; Tarling, 1981; Hutchison, 1983; Sawkins, 1990; Groves and Bierlein, 2007). The following sections
provide an overview of the major plate-related tectonic settings and the ore deposit types associated with each.

Extensional Settings
Incipient rifting of stable continental crust is represented in Fig. 10A, where thinning and extension may also involve plume activity.
The resulting alkaline or ultrapotassic magmatism (to form kimberlites and lamproites) is often localized along craton margins or
intracratonic rifts, and the products preserved along lineaments or sutures. Alkaline granites such as those of the Bushveld Complex
(Sn, W, Mo, Cu, F, etc.), alkaline-carbonatite intrusions such as Phalaborwa (Cu-Fe-P-U-REE) in South Africa and Mount Weld
(REE-Y-Nb-Ta-P-Zr) in Western Australia, as well as kimberlites (diamonds), represent ore deposit types formed in this setting.
Intracontinental rifts can also host SEDEX-type Pb-Zn-Ba-Ag deposits as well as stratiform sediment-hosted Cu-Co deposits.
As continental rifting extends to the point that incipient oceans begin to open (such as the Red Sea; Fig. 10B), basaltic volcanism
accompanies the formation of a new mid-ocean ridge along which exhalative hydrothermal activity and VMS deposit formation
might occur. Such settings provide the environments for chemical sedimentation and precipitation of banded iron-formations and
manganiferous sediments. Continental platforms commonly host organic accumulations that on catagenesis give rise to oil
deposits. Carbonate sedimentation ultimately provides the rocks which host MVT deposits, although the hydrothermal processes
that give rise to these epigenetic Pb-Zn ores are typically associated with circulation during compressional stages of orogeny. Mid-
ocean ridges are the culmination of extensional processes (Fig. 10C). Exhalative activity at these sites gives rise to “black-smoker”
vents (such as those at 21 N on the East Pacific Rise) that provide the environments for the formation of Cyprus type VMS deposits.
The basalts that form at mid-ocean ridges undergo fractional crystallization at sub-volcanic depths and may form podiform
chromite deposits as well as Cu-Ni-PGE sulfide segregations.
 Plate Tectonics and Metallogeny 17

Fig. 10 Simplified illustrations of the major extensional tectonic settings and the ore deposit types associated with each (modified after Mitchell and Garson,
1981).

Compressional Settings
Andean type (ocean-continent) collisional margins are both wide-spread and important as metallotects (Fig. 11). These are the sites
of the porphyry Cu-Mo provinces of the world, while significant Sn-W granitoid-hosted mineralization also forms inboard of the
arc. The volcanic regions above the porphyry systems are the sites of epithermal precious metal mineralization. A similar tectonic
setting can exist between two colliding slabs of oceanic crust, as represented by the island arc environment in Fig. 11B. Porphyry
18 Plate Tectonics and Metallogeny

Fig. 11 Simplified illustrations of the major compressional tectonic settings and the ore deposit types associated with each (modified after Mitchell and Garson,
1981).
 Plate Tectonics and Metallogeny 19

Cu-Au deposits occasionally occur associated with the early stages of magmatism in these settings, whereas the later, more evolved
calc-alkaline magmatism gives rise to Kuroko-type VMS deposits. Back-arc basins represent the sites of Besshi type VMS deposition.
Arc–arc collision in the back-arc environment can also result in the preservation of obducted oceanic spreading centers within which
podiform Cr and sulfide segregations might be preserved. In a Japanese-style setting, where the island arc develops fairly close to a
continent (Fig. 11C), marginal sedimentary basins floored by oceanic crust occur. This setting typically hosts Besshi- and Cyprus-
type VMS deposits. As the arc and continent accrete, ophiolite obduction can occur, and felsic magmatism may give rise to large-ion
lithophile element mineralization. Ultimately, the oceanic crust is totally consumed to form a zone of continent-continent collision
(Fig. 11D). Modern examples such as the Himalayas and Alps do not appear to be significantly mineralized, but this may be an
expression of insufficient exhumation of mineralized zones. Older examples preserve Sn-W-U mineralization in S-type granites,
whereas orogeny-driven fluids give rise to orogenic vein-related lode Au systems and MVT Pb-Zn deposits in suitably preserved
platformal sediments.

Acknowledgments

LR acknowledges the support of the DSI-NRF Centre of Excellence for Integrated Mineral and Energy Resource Analysis (DSI-NRF
CIMERA) towards this research. CH acknowledges the support of the Leverhulme Trust through grants RPG-2015-422 and
EM-2017-047\4. We thank Dave Sansom for drafting the diagrams.

References

Abbott DH and Isley AE (2002) The intensity, occurrence and duration of superplume events and eras over geological time. Journal of Geodynamics 34: 265–307.
Appel PWU (1983) Mineral occurrences in the 3.6 Ga old Isua supracrustal belt, west Greenland. In: Trendall AF and Morris RC (eds.) Iron Formation: Facts and Problems.
Developments in Precambrian Geology, vol. 6, pp. 593–603. Elsevier.
Armstrong RL (1981) Radiogenic isotopes: The case for crustal recycling on a near-steady-state no-continental growth earth. Philosophical Transactions of the Royal Society
A301: 443–472.
Barley ME and Groves DI (1992) Supercontinent cycles and distribution of metal deposits through time. Geology 20: 291–294.
Barley ME, Bekker A, and Krapez B (2005) Late Archean to Early Proterozoic global tectonics, environmental change and the rise of atmospheric oxygen. Earth and Planetary Science
Letters 238: 156–171.
Bédard JH (2018) Stagnant lids and mantle overturns: Implications for Archaean tectonics, magmagenesis, crustal growth, mantle evolution and the start of plate tectonics.
Geoscience Frontiers 9: 19–49.
Bekker A, Holland HD, Wang P-L, Rumble D III, Stein HJ, Hannah JL, Coetzee LL, and Beukes NJ (2004) Dating the rise of atmospheric oxygen. Nature 427: 117–120.
Bekker A, Slack JF, Planavsky N, Krapez B, Hofmann A, Konhauser K, and Rouxel OJ (2010) Iron formation: The sedimentary product of a complex interplay among mantle, tectonic,
oceanic and biospheric processes. Economic Geology 105: 467–509.
Bekker A, Planavsky N, Krapez B, Rasmussen B, Hofmann A, Slack JF, Rouxel OJ, and Konhauser K (2014) Iron formations: Their origins and implications for ancient seawater
chemistry. In: Holland HD and Turekian KK (eds.) Treatise on Geochemistry. 2nd edn, vol. 13, pp. 561–628. Oxford: Elsevier.
Belousova EA, Kostitsyn YA, Griffin WL, Begg GC, O’Reilly SY, and Pearson NJ (2010) The growth of the continental crust: Constraints from zircon Hf-isotope data. Lithos
119: 457–466.
Bleeker W (2003) The late Archean record: A puzzle in ca. 35 pieces. Lithos 71: 99–134.
Bradley DC (2011) Secular trends in the geological record and the supercontinent cycle. Earth-Science Reviews 108: 16–33.
Button A (1979) Transvaal and Hamersley basins—Review of basin development and mineral deposits. Minerals Science and Engineering 8(4): 262–293.
Campbell IH (2003) Constraints on continental growth models from Nb/U ratios in the 3.5 Ga Barberton and other Archaean basalt-komatiite suites. American Journal of Science
303: 319–351.
Cawood PA and Hawkesworth CJ (2015) Temporal relations between mineral deposits and global tectonic cycles. In: Jenkin GRT, Lusty PAJ, McDonald I, Smith MP, Boyce AJ, and
Wilkinson JJ (eds.) Ore Deposits in an Evolving Earth, Geological Society, London, Special Publication, vol. 339, pp. 9–21.
Cawood PA, Hawkesworth CJ, Pisarevsky SA, Dhuime B, Capitanio FA, and Nebel O (2018) Geological archive of the onset of plate tectonics. Philosophical Transactions of the Royal
Society A376. https://doi.org/10.1098/rsta.2017.0405.
Chalapathi Rao NV and Lehmann B (2011) Kimberlites, flood basalts and mantle plumes: New insights from the Deccan Large Igneous Province. Earth-Science Reviews
107: 315–324.
Charlier B, Namur O, Bolle O, Latypov R, and Duchesne JC (2015) Fe-Ti-V-P ore deposits associated with Proterozoic massif-type anorthosites and related rocks. Earth-Science
Reviews 141: 56–81.
Cheney ES (1996) Sequence stratigraphy and plate tectonic significance of the Transvaal succession of southern Africa and its equivalent in Western Australia. Precambrian Research
79: 3–24.
Cook PJ and Shergold JH (1984) Phosphorus, phosphorites and skeletal evolution at the Precambrian-Cambrian boundary. Nature 308: 2331–2337.
Denison RE, Lidiak EG, Bickford ME, and Kisvarsanyi EB (1984) Geology and geochemistry of the Precambrian rocks in the Central Interior Region of the United States. In: Geological
Survey Professional Paper 1241-C, 20pp.
Dhuime B, Hawkesworth CJ, Cawood PA, and Storey CD (2012) A change in the geodynamics of continental growth 3 billion years ago. Science 335: 1334–1336.
Evans DAD and Mitchell RN (2011) Assembly and break-up of the core of Paleoproterozoic-Mesoproterozoic supercontinent Nuna. Geology 39(5): 443–446.
Filippelli GM (2008) The global phosphorus cycle: Past, present, and future. Elements 4: 89–95.
Fyfe WS (1978) The evolution of the earth’s crust: Modern plate tectonics to ancient hot spot tectonics? Chemical Geology 23: 89–114.
Greyling LN, Robb LJ, Master S, and Boiron MC (2005) The nature of early mineralising fluids associated with the Chambishi deposit, Zambian Copperbelt. Journal of African Earth
Sciences 42: 159–172. Special Issue.
Groves DI and Bierlein FP (2007) Geodynamic settings of mineral deposit systems. Journal of the Geological Society of London 164: 19–30.
Hallam A (1992) Phanerozoic Sea-Level Changes, 266pp. Columbia University Press.
Harland WB (1965) Critical evidence for a great infra-Cambrian glaciation. Geologische Rundschau 54: 45–61.
Hawkesworth CJ, Dhuime B, Pietranik AB, Cawood PA, Kemp AIS, and Storey CD (2010) The generation and evolution of the continental crust. Journal of the Geological Society of
London 167: 229–248.
20 Plate Tectonics and Metallogeny

Hawkesworth CJ, Cawood PA, and Dhuime B (2016) Tectonics and crustal evolution. GSA Today 26(9): 4–11.
Herzberg C, Condie K, and Korenaga J (2010) Thermal history of the earth and its petrological expression. Earth and Planetary Science Letters 292: 79–88.
Hoffman PF (1988) United plates of America, the birth of a craton: Early Proterozoic assembly and growth of Laurentia. Annual Review of Earth and Planetary Sciences 16: 543–603.
Hoffman PF, Kaufman AJ, Halverson GP, and Schrag DP (1998) A Neoproterozoicsnowball Earth. Science 281: 1342–1346.
Holland HD (2006) The oxygenation of the atmosphere and oceans. Philosophical Transactions of the Royal Society B361: 903–915.
Hutchison CS (1983) Economic deposits and their tectonic setting, 365pp. Macmillan Press.
Johnson JP and Cross KC (1995) U-Pb geochronological constraints on the genesis of the Olympic Dam Cu-U-Au-Ag deposit, South Australia. Economic Geology 90(5): 1046–1063.
Kasting JF (1993) Evolution of the Earth’s atmosphere and hydrosphere. In: Engel MH and Macko SA (eds.) Organic Geochemistry, 821pp. Plenum Press.
Korenaga J (2013) Initiation and evolution of plate tectonics on earth: Theories and observations. Annual Review of Earth and Planetary Sciences 41: 117–151.
Larson RL (1991) Geological consequences ofsuperplumes. Geology 19: 963–966.
Martin DM, Clendenin CW, Krapez B, and McNaughton NJ (1998) Tectonic and geochronological constraints on late Archaean and Palaeoproterozoic stratigraphic correlation within
and between the Kaapvaal and Pilbara cratons. Journal of the Geological Society of London 155: 311–322.
Master S, Rainaud C, Armstrong RA, Phillips D, and Robb LJ (2005) Provenance ages of the Neoproterozoic Katanga Supergroup (Central African Copperbelt), with implications for
basin evolution. Journal of African Earth Sciences 42: 41–60.
McCulloch MT and Bennett VC (1994) Progressive growth of the Earth’s continental crust and depleted mantle. Geochemical constraints. Geochimica et Cosmochimica Acta
58: 4717–4738.
McLennan SM and Taylor RS (1982) Geochemical constraints on the growth of the continental crust. Journal of Geology 90: 347–361.
McMenamin MAS and McMenamin DLS (1990) The emergence of animals: The Cambrian breakthrough, 217pp. New York: Columbia University Press.
Meyer C (1981) Ore-forming processes in geologic history. In: Economic Geology, 75th Anniversary Volume, pp. 6–41.
Meyer C (1988) Ore deposits as guides to the geologic history of the Earth. Annual Review of Earth and Planetary Sciences 16: 147–171.
Mitchell AHG and Garson MS (1981) Mineral Deposits and Global Tectonic Settings, 457pp. Academic Press.
Mojzsis SJ (2001) Life and the evolution of Earth’s atmosphere. In: Mathez E (ed.) Earth Inside and Out, pp. 32–39. New York: New Press.
Mosier DL, Berger VI, and Singer DA (2009) Volcanogenic Massive Sulfide Deposits of the World—Database and Grade and Tonnage Models. US Geological Survey, Open File Report,
2009–1034.
Nance RD and Murphy JB (2013) Origins of the supercontinent cycle. Geoscience Frontiers 4: 439–448.
Nance RD, Worsley TR, and Moody JB (1986) Post-Archean biogeochemical cycles and long-term episodicity in tectonic processes. Geology 14: 514–518.
Nance RD, Murphy JB, and Santosh M (2014) The supercontinent cycle: A retrospective essay. Gondwana Research 25: 4–29.
Park RG (1995) Palaeoproterozoic Laurentia-Baltica relationships: A view from the Lewisian. In: Coward MP and Ries AC (eds.) Early Precambrian Processes, 95, pp. 211–224.
Geological Society, Special Publication.
Piper JD (1976) Palaeomagnetic evidence for a Proterozoic super-continent. In: Sutton J, Shackleton RM, and Briden JC (eds.) A Discussion on Global Tectonics in Proterozoic Times,
Philosophical Transactions of the Royal Society of London 280: 469–490.
Rogers JJW (1996) A history of continents in the past three billion years. Journal of Geology 104: 91–107.
Rogers JJ and Santosh M (2004) Continents and supercontinents, 289pp. Oxford University Press.
Sawkins FJ (1990) Metal deposits in relation to plate tectonics, 461pp. Springer-Verlag.
Scotese CR (2009) Late Proterozoic plate tectonics and paleogeography: A tale of two supercontinents, Rodinia and Pannotia. Geological Society of London, Special Publication
326: 67–83.
Shirey SB and Richardson SH (2011) Start of the Wilson cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333: 434–436.
Tarling DH (1981) Economic Geology and Geotectonics, 213pp. Blackwell Scientific.
Titley SR (1993) Relationship of stratabound ores with tectonic cycles of the Phanerozoic and Proterozoic. Precambrian Research 61: 295–322.
Torsvik TH and Cocks LRM (2013) Gondwana from top to base in space and time. Gondwana Research 24: 999–1030.
Veizer J, Laznicka P, and Jansen SL (1989) Mineralization through geologic time: Recycling perspective. American Journal of Science 289: 484–524.
Volp KM (2005) The Estrela copper deposit, Carajas, Brazil: Geology and implications of a Proterozoic copper stockwork. In: Mineral Deposit Research—Meeting the Global Challenge,
pp. 1085–1088. Berlin: Springer.
Williams H, Hoffman PH, Lewry JF, Monger JWH, and Rivers T (1991) Anatomy of North America: Thematic geologic portrayals of the continents. Tectonophysics 187: 117–134.
Windley BF (1995) The evolving continents, 526pp. Wiley.
Worsley TR, Nance D, and Moody JB (1984) Global tectonics and eustacy for the past 2 billion years. Marine Geology 58: 373–400.
Zhang S, Li Z-X, Evans DAD, Wu H, Li H, and Dong J (2012) Pre-Rodinia supercontinent Nuna shaping up: A global synthesis with new paleomagnetic results from North China. Earth
and Planetary Science Letters 353-354: 145–155.
Zhao G, Sun M, Wilde S, and Li SZ (2004) A paleo-Mesoproterozoic supercontinent: Assembly, growth and break-up. Earth-Science Reviews 67: 91–123.

Further Reading
Bradley DC (2011) Secular trends in the geological record and the supercontinent cycle. Earth-Science Reviews 108: 16–33.
Cawood PA and Hawkesworth CJ (2015) Temporal relations between mineral deposits and global tectonic cycles. In: Jenkin GRT, Lusty PAJ, McDonald I, Smith MP, Boyce AJ, and
Wilkinson JJ (eds.) Ore Deposits in an Evolving Earth, Geological Society, London, Special Publication, vol. 339, pp. 9–21.
Goldfarb RJ, Groves DI, and Gardoll S (2001) Orogenic gold and geologic time: A global synthesis. Ore Geology Reviews 18: 1–75.
Groves DI and Bierlein FP (2007) Geodynamic settings of mineral deposit systems. Journal of the Geological Society of London 164: 19–30.
Hawkesworth CJ, Dhuime B, Pietranik AB, Cawood PA, Kemp AIS, and Storey CD (2010) The generation and evolution of the continental crust. Journal of the Geological Society of
London 167: 229–248.
Hawkesworth CJ, Dhuime B, and Cawood PA (2016) Tectonics and crustal evolution. GSA Today 26(9): 4–11. https://doi.org/10.1130/GSATG272A.1.
Hutchison CS (1983) Economic Deposits and Their Tectonic Setting, 365pp. London: Macmillan Press.
Kesler S (1997) Metallogenic evolution of convergent margins: Selected ore deposit models. Ore Geology Reviews 12: 153–171.
Sawkins FJ (1990) Metal Deposits in Relation to Plate Tectonics, 2nd edn., 461pp. New York: Springer-Verlag.
Windley BF (1995) The Evolving Continents, 526pp. New York: John Wiley and Sons.