The earth comprises a number of lithospheric plates that move apart at mid-oceanic ridges, are consumed at subduction zones, collide with each other in collisional orogens, or slide past each other along transform boundaries. Although oceanic crust is continually being created and destroyed, long-lived stable parts of continents called cratons have remained undeformed for billions of years. Continental plates containing ancient cratons have episodically collided and assembled in global periods of orogenesis to form supercontinents. Supercontinents eventually become unstable, as such a large single landmass acts as a thermal lid, limiting escape of Earth's internal heat. Supercontinent breakup occurs when old crustal weaknesses (such as orogenic belts created during supercontinent assembly) overlay several mantle plumes , or due to the formation of a superplume. Dispersed fragments move across the globe to subsequently amalgamate to form another supercontinent. The process of supercontinent formation, breakup, and dispersal has continued cyclically through Earth's history.
Plate reconstructions for periods younger than 180 million years (the oldest age of crust in present-day oceans ) can be made by graphically undoing seafloor spreading magnetic anomalies of known age. Offshore gravity , calculated from satellite altimetery of the sea surface, defines continental margins and structures in oceanic crust that can also be used in recent continental reconstructions. Rotation poles (Euler poles), about which lithospheric plates are displaced, can be determined from transform faults interpreted from satellite altimeter-derived images. All reconstructions are based on the generally held view that Earth has maintained a constant radius (although this is questioned by some who contend that the earth has progressively expanded). For older periods, scientists rely on the following to establish, or at least infer, continental correlations:
- Linking of orogenic belts and intracratonic structures (e.g., major shear zones of the same age and displacement sense)—accurate dating is unfortunately lacking from many Precambrian terrains, hindering such correlations. Regional aeromagnetic data is valuable in comparing lithological trends and structural elements as basement rocks below superficial cover and sedimentary basins are imaged. Regional gravity data highlights major crustal structures.
- Mafic dikes—dike swarms (readily discernable on aero-magnetic images) may be traced from one craton to another. Dikes may converge on an area above the center of a former mantle plume.
- Common rock types, ages, and fossil assemblages in sedimentary rocks on conjugate margins—provenance studies and ages of detrital zircons for sedimentary basins provide information concerning the source terrains for sedimentary basins. Sedimentary source rocks absent from the craton the basin is situated in may be found on previously contiguous continents.
- Paleomagnetism and polar wander paths—when sedimentary rocks are deposited or when igneous or metamorphic rocks cool below 578° for magnetite and 675° for hematite (the minerals' Curie temperatures), these iron-rich minerals preserve the orientation of the earth's magnetic field . Rocks of the same age from joined continents exhibit a common magnetic pole. Apparent polar wander paths linking poles of different ages graphically portray the displacement of a continent with time. Changes in previously overlapping apparent polar wander paths for two continents indicate the time continents rifted apart.
- Large igneous provinces of the same age and with characteristic geochemical signatures (representing portions of the same igneous province prior to breakup)— Archean to late Mesoproterozoic correlations are, however, highly problematic due to the poor to non-existent paleomagnetic database, the wide dispersal of cratons following their breakup and the likely disruption or tilting of old cratons during reworking along craton margins.
An Archean , 3.1 (to possibly even 3.6) billion years old supercontinent Vaalbara, in which the South African Kaapvaal Craton was joined to the Pilbara Craton of Western Australia , is the oldest proposed supercontinent. The concept of Vaalbara is based on similarities in sedimentary sequences on both cratons. The existence of Vaalbara has, however, been questioned due to differences in magmatic events between both cratons, and the possibility for similar sedimentary successions to have been deposited on separate continents due to global sea level changes. Recent paleomagnetic evidence suggests that the cratons were not contiguous about 2.8 billion years ago.
Another Archean to Paleoproterozoic supercontinent, Ur (the German word for original) has been proposed. Ur is thought to comprise a nucleus of the Kaapvaal and Pilbara cratons (although not adjacent to each other as in Vaalbara), the Indian Bhandara and Singhbhum cratons, and possibly some Archean East Antarctic terrains. A 1.8 to 1.5 billion year supercontinent Columbia comprising most continents then existing on Earth has also been proposed in which eastern India, Australia, and their contiguous portion of East Antarctica was joined to western North America .
Rodinia (from the Russian rodit, meaning to grow) is the late Mesoproterozoic to late Neoproterozoic (approximately 1,200 to 700 million year) supercontinent formed by the assembly of Precambrian terrains of Australia, North America-Canada, India, Madagscar, Sri Lanka, and East Antarctica. The idea for such a supercontinent initially came from the recognition of geological similarities between western Canada and southeastern Australia, and likely links between 1.3 to 0.9 billion-year-old orogenic belts. The term Grenvillian belts has been loosely used to encompass orogens formed during this broad time span. This is, however, not the strict definition of the Grenvillian Orogeny , which is defined as occurring between 1,090 and 980 million years ago in its type area, the Grenville Province of North America. There is still much debate as to the configuration of continents within Rodinia.
In the first SWEAT (from S outhw estern United States–E ast A nt arctica) configuration, Laurentia is positioned such that western Canada is opposite eastern Australia. Grenvillian orogens (such as the Central Indian Tectonic Zone and Pinjarra Orogen of Western Australia) that were not part of a continuous orogenic belt were omitted in this reconstruction, giving it a somewhat false simplicity. In a modified version, South China (formed through collision of the Cathaysia and Yangtze blocks) is positioned between Australia and Laurentia. In another reconstruction for Rodinia, AUSWUS (Aus tralia–W estern U nited S tates), eastern Australia is attached to the western United States. Both are compatible with available paleomagnetic data for 1,140 million years; however, both are not compatible with recent paleomagnetic poles for 1,070 million-year-old rocks from Laurentia and Western Australia. In an attempt to explain these poles, a reconstruction of Rodinia called AUSMEX (Aus tralia– Mex ico) has Laurentia in a rotated position with respect to Australia, with the Grenville Province of North America continuing directly into northeastern and central Australia. In the SWEAT and AUSWUS reconstructions, northeastern India joins with southwestern Australia, and southeastern India is linked to Antarctica. Despite sound geological links between Proterozoic orogenic belts supporting this configuration, current paleomagnetic data suggest that India and part of East Antarctica may not have amalgamated with Australia until 680 to 610 million years ago. Clearly, there are still many problems to be resolved before consensus is reached for a reconstruction of Rodinia compatible with all available data. It must also be asked whether palaeomagnetic data alone is a reliable means of establishing ancient positions of continents.
Rodinia split into two main fragments approximately 750 million years ago. Pannotia (meaning all southern) is the name given to a supercontinent formed when the northern block (comprising Australia, Antarctica, India, Madagascar, Arabia, and parts of China), Laurentia, and cratonic blocks in East Africa , Mozambique, Madagascar, and South America were amalgamated through collision in the Pan-African orogeny. Pannotia broke up into Laurentia, Baltica, Siberia, and Gondwanaland at the end of the Precambrian, about 590 million years ago.
The idea of links between India, central and southern Africa, and Madagascar based on the common occurrence of glossopteris (a fossilized seed fern first described in early Permian coal seams in central India) and other fossil assemblages was first proposed in 1885 by the Austrian geologist Eduard Suess. Suess coined the term Gondwanaland, meaning the land of the kingdom of the Gons (a Dravidian people in central India) for the area he thought to have been linked by land bridges between fixed continents. Australia and Antarctica were subsequently added following further glossopteris discoveries. The proposition of links between continents predated the concepts of continental drift and plate tectonics , but was subsequently used as evidence in their formulation. Gondwanaland continents contain similar Permo-Carboniferous (286 million year old) to late Jurassic–early Cretaceous (100 million year) sedimentary successions. Permo-Carbonifereous tillites and other similar glacial deposits in India, Australia, and Antarctica are the vestiges of a large ice sheet. Older ice sheets were present in Africa and South America. Early reconstructions of Gondwanaland were based on the outlines of continents, structural features, and broad matches in geology . These have been subsequently refined using paleomagnetic and seafloor spreading data.
Gondwanaland was separated by the Tethys Ocean from another supercontinent, Laurasia, formed through collision of Laurentia (North America), Baltica (Scandinavian continents), parts of Europe , and Siberia approximately 400 million years ago. The supercontinent Pangaea (meaning all Earth) was formed by the collision of Laurasia with Gondwanaland approximately 275 million years ago following closure of part of the Tethys, and the collision with Cimmeria (fragments of Turkey, Afghanistan, Iran, Tibet, and Indochina). Mountain belts such as the Appalachians and Urals were formed in this event. The breakup of Pangaea in the Late Jurassic to Early Cretaceous occurred largely due to rifting along old weaknesses when they became aligned between mantle plumes.
Understanding the configurations of past supercontinents is of more than academic interest. The formation and dispersal of supercontinents has had a marked effect on past changes in ocean circulation patterns and hence on Earth's climate . Major mineral provinces on one continent may have as yet undiscovered corollaries on another, once adjacent continent. Placer gold deposits in a sedimentary basin on one continent may have been eroded from vein or shear zone hosted deposits on another continent. The Mt. Isa Belt in Queensland, Australia, is truncated by a rifted margin formed during the breakup of Rodinia. Rich gold and base metal deposits may exist in its continuation on another, yet undefined terrain, possibly within southeast China or North America (placed against Queensland in various Rodinia reconstructions).
See also Continental drift theory; Earth, interior structure; Earth (planet); Geologic time