Rocks and Plate Tectonics

Steven Dutch, Professor Emeritus, Natural and Applied Sciences, Universityof Wisconsin - Green Bay

General Remarks

Plate tectonics strongly governs the rocks that occur in specific settings, either by the physical processes of plate tectonics themselves, which govern igneous and metamorophic rocks, or by creating topography that governs sedimentary environments.

Stable Continental Interiors

Stable continental interiors are, well, stable. They are dominated by erosion, so sedimentary rocks will be preserved only if they were deposited in subsiding basins or buried by later rocks. Volcanism and intrusive activity are rare and confined to occasional dikes and cinder cones. Deeply eroded areas may expose ancient shield rocks which may be igneous or metamorphic. Principal rocks include:

Terrestrial and fresh-water rocks like mature quartz sandstones, shales, coal, and occasional evaporites. Some of the sandstones, like the Navajo Sandstone of the Southwest, are dune sands. In fact, it's the largest dune field in the geologic record, past or present. Distinctive fossils include leaves, wood fragments, insects and terrestrial vertebrates. Insects and amphibians are almost exclusively terrestrial or fresh water organisms.
During periods of marine transgression, large areas of the continents may be covered by shallow marine rocks, which may include sandstone, shale, limestone and dolomite (derived by alteration of limestone). Some evaporites may form in restricted basins. Distinctive fossils include brachiopods, molluscs, bryozoans, echinoderms (crinoids, starfish, sea urchins) corals and fish. Brachiopods, cephalopods, bryozoans, echinoderms and corals are exclusively marine (there are fresh-water bryozoans but never as abundant fossils).

Passive Continental Margins

Passive continental margins are those where continents have rifted apart and the rocks after rifting are entirely derived by erosion of the continent. The continental margins around the Atlantic and Indian Oceans are mostly of this type.

The initial rifting is characterized by faulting and volcanism. Basalt lava flows and diorite or gabbro dikes and sills are common. Since subsidence is rapid, sedimentary rocks are predominately coarse terrestrial clastic rocks. There may be evaporites in some closed rift basins. The Newark Basin of New Jersey is the classic example.
On the continental shelf, later rocks are predominately shallow marine rocks thickening seaward. They are principally sandstone and shale, with some carbonates.
If warm conditions exist, the continental shelf may be dominated by carbonate rocks, with a barrier reef along the edge of the shelf. Australia, Belize, Yucatan and the Florida Keys are contemporary examples. Western Newfoundland contains a spectacular talus conglomerate formed when the North American continental margin was bounded by a reef complex during the Ordovician.
Beyond the edge of the shelf, sediments are predominately poorly sorted sandstones and shales. The water is too deep for carbonate-depositing life, and once sediment is deposited this deep, opportunities for weathering and reworking are nil. What you get is what you get. Since the sediment is resting on a slope, slumps and submarine landslides are common, producing turbidites.
Continental Rifts and Hot Spots
Rapid continental rifting and vigorous hot spots both result in voluminous basalt flows. Typical examples of hot spots include the Siberian Traps of Permian age, the Cretaceous Deccan Traps of India and the Miocene Columbia Plateau basalts of Washington and Oregon. Rift examples include the Parana volcanic rocks of Brazil and the Karroo volcanic rocks of Africa, related to opening of the South Atlantic.
Slower or more protracted rifting and hot spot activity can create a bimodal set of volcanic rocks with copious early basalt, later rhyolite, and virtually no intermediate rocks. This pattern is probably caused by heating and melting (anatexis) of continental crust to produce the silica rich lavas. Examples include the Basin and Range Province of the western U.S., the Yellowstone hot spot, and the Precambrian Mid-Continent Rift of the Great Lakes region.
Very slow or abortive rifting, and weak hot spot activity are often marked by volcanic rocks of extreme composition. Possibly volatiles and incompatible elements in the long-undisturbed mantle beneath the continent are the first ingredients to mobilize when presented with an opportunity to escape. Examples include the Ottawa River Graben in Canada, which incudes diatremes, carbonatites, alkalic intrusions and fenites (breccia pipes with highly sodic alteration). The East African Rift is famed for Ol Doinyo Lengai, a volcano that erupts sodium carbonate lava.

Ocean Basins

The basement, or underlying rock of the ocean basins, is basalt with gabbro intrusions. These rocks are exposed along the mid ocean ridges and sometimes where faults have exposed buried ocean floor. Hot fluids escaping from the mid ocean ridge can create deposits of copper, zinc and other minerals. Much of the basalt is pillow lava, a distinctive texture that results from sudden cooling of lava by sea water.
Much of the ocean floor is covered by the skeletons of silica-secreting microorganisms called radiolaria. This material is called radiolarian ooze if it's unconsolidated, or radiolarian chert once it's lithified. All this material formed in the photic zone where sunlight sustains the food chain.
Large areas of the ocean floor are covered with red clay, believed to be derived from chemical precipitates, aeolian continental material, and meteorite debris. Manganese nodules, probably precipitated biogenically, cover parts of the ocean floor.
Carbonates are limited by the Carbonate Compensation depth, the depth below which calcium carbonate becomes soluble under pressure, or about 4-5 km. Carbonates already deposited dissolve slowly and can persist below that depth, but microscopic particles dissolve on the way down. Carbonate sediments on the ocean floor can be a proxy for past changes in oceanic carbon dioxide content and vertical motions of sea water. Carbonate sediments are mostly the remains of single celled animals called foraminifera and algae called coccolithophores.
Aeolian continental sediments are mostly indicated by the presence of quartz or kaolinite.
Glacio-marine sediments include fine clays carried in suspension, and coarser material carried by floating ice. Pretty much the only way sand can get to the middle of an ocean is by attachment to something afloat.
Turbidites originating at the edge of the continental shelf and on the continental slope can travel for long distances across the flat abyssal plains. They can be recognized by relative coarseness, graded bedding, and distinctive scour features on the bottom of the bed.
Ocean basins can include large submarine volcanic plateaus and hot spot island chains like Hawaii. These are almost entirely basalt, but very long lived magma chambers can differentiate enough to produce small amounts of more silica rich rocks. Coral reefs may build up in shallow areas.

Subduction Zones

The principal types of rocks found in subduction zones are pre-existing rocks, modified by deformation and metamorphism, sediment eroded off the overriding plate, oceanic materials derived from the descending plate, and igneous rocks derived by melting of the mantle wedge between the two converging plates.

If the two converging plates are both oceanic crust (like the Philippine and Pacific Plates), then ultimately, pretty much everything is derived from basalt. The volcanic rocks are basalt, intrusions are gabbro, and sedimentary rocks are derived from weathering and erosion of basalt. There are minor contributions from ocean floor material and coral reefs, and some modification of the sediments occurs during weathering. Over time, a fairly large land mass can be created (Panama, Costa Rica, Cuba) and repeated subduction and re-melting of rocks can even result in granite. Still, lithologically, Panama and Costa Rica are pretty monotonous.
During subduction, material can be scraped off the descending oceanic plate and onto the overriding plate. One of the most distinctive materials is thin-bedded radiolarian chert, a common rock type in the California Coast Ranges.
Sometimes larger fragments of oceanic crust break off and are pushed (obducted) onto the overriding plate. These can include seamounts and slivers of oceanic crust, or ophiolites. Pillow lavas are distinctive features.
Well preserved ophiolites like those in Cyprus or Oman are valuable sources of information about ocean floor processes, but many others are hydrothermally altered to serpentine and contain few obvious structural clues to their origin. Most serpentine belts in mountain ranges are now believed to be altered ophiolites.
Often the pillow lavas in ophiolites are sodium rich and are termed spilites. Spilites were once a deep mystery but they are now recognized as the products of simple ion exchange where calcium in feldspars was replaced by sodium from sea water. The rhyolite equivalents are termed keratophyres.
Long before plate tectonics was accepted, geologists had noted the common association of radiolarian cherts, pillow lavas and serpentine. This grouping is called the Steinmann Trinity.
Sediment is eroded off the overriding plate and dumped without ceremony into the oceanic trench. With little weathering during transport, the resulting rock is extremely immature and poorly sorted. The rock is called greywacke, and the rock suite is called flysch. Sometimes the flysch is chaotically deformed by tectonism and slumping and is termed Wildflysch (German, and pronounced vild-flysch. It means just what it looks like).
Pre-existing rocks on the overriding plate can be any type whatever and offer clues to the history of the plate before the subduction zone developed.
Rocks along the subduction zone are frequently extremely deformed, tectonically disrupted, and intermixed. Such an assemblage is called a melange (French: mixture). The California Coast Ranges are basically a giant melange zone. Geologists went batty for a century trying to sort out the Coast Ranges before realizing they can't be sorted out.
Rocks along the subduction zone are frequently carried to depths of 40 or more kilometers, but contact with the cool descending slab keeps them from being heated to very high temperatures. The result is a highly distinctive type of metamorphism called blueschist metamorphism, so called because many of the characteristic minerals are blue. Blueschist metamorphism is the smoking gun for a former subduction zone. How rocks are exhumed from such great depths so quickly is a bit of a mystery. Occasionally there will be pods of ultra-deep metamorphic rocks called eclogite as well.
For reasons not fully understood, blueschists are rare in rocks older than Mesozoic. In very old rocks, the geothermal gradient in the earth may have been too high for blueschists. By the time they got to sufficient depths, temperatures were in the greenschist or amphibolite range. It may also be that many blueschists are subsequently re-metamorphosed to more conventional greenschists.

Active Mountain Belts (Orogenic Belts)

Much of what happens in active mountain belts inboard from the subduction zone proper is related to the igneous arc, the zone of maximum volcanism and intrusion. Although it's common to read in textbooks about the "melting of the descending slab," the descending slab only melts if the oceanic crust is very young and still quite hot. What actually happens in the vast majority of cases is that water is cooked off the descending slab and lowers the melting point of the mantle between the converging slabs (the so called "mantle wedge"). The mantle wedge partially melts, producing a magma that is more silica rich than the original mantle.

Volcanism in orogenic belts begins as mostly basaltic and becomes more silica rich over time, with abundant andesite and finally rhyolite.
Intrusions in orogenic belts can be initially gabbro or tonalite in composition. Tonalite consists mostly of plagioclase, quartz and ferromagnesian minerals with very little potassium feldspar. Later, the largest intrusions are often batholiths of granodiorite, and later, granite.
The progression from basalt through andesite to rhyolite, and from tonalite through granodiorite to granite, both reflect the increasing differentiation of magma over time, as well as the increasing mixing of magma with pre-existing continental rocks.
Metamorphism is most intense in the igneous arc, with amphibolite facies metamorphism in the core of the orogenic belt and greenschist metamorphism at higher levels and around the periphery.
An ideal orogenic belt would have paired metamorphic belts, with blueschist metamorphism marking the subduction zone and greenschist-amphibolite metamorphism along the igneous arc. Most orogenic belts deviate from this ideal by incorporating terranes (note the spelling), which are blocks of crust that originated elsewhere and were carried into the subduction zone by plate convergence. Terranes can be volcanic island chains, pieces of continental crust, or submarine volcanic plateaus. Obviously, anything goes when it comes to lithology of exotic terranes. Virtually always, terranes are bounded by faults.
Apart from places where blocks of crust are pushed upward along thrust faults, orogenies do not uplift mountains directly. Mostly, orogenic processes thicken the crust by compression or by adding material, and isostasy cause the crust to rise.
When continents collide, generally the former plate boundary is marked by a suture zone with blueschists, serpentine, and major thust faults. The overriding plate will usually show signs of active orogeny while the continent that was underthrust will not. India and Tibet are premier examples.
Along continental collision boundaries, the overriding continent may be thrust upward to expose deep crustal rocks. One of the best examples in the world is the Ivrea Zone in the Alps of northwestern Italy, where a cross section of the crust is exposed, including extremely deep dehydrated rocks called granulites. In granulites, all the hydrous minerals have broken down and the only ferromagnesian minerals are pyroxenes.
The most extreme metamorphic rocks are probably gneisses found in China, which were estimated to have reached depths of 100 kilometers. These gneisses appear to have contained coesite, a high pressure polymorph of silica. The coesite reverted to quartz at lower pressures, and in the process expanded by about 1/3. The former coesite grains are recognized by distinctive fracture patterns within and around the grains due to the expansion.
Once mountains have been uplifted, erosion creates great quantities of coarse terrestrial sediment that buries the flanks of the mountains. Much of central Switzerland is covered by kilometers of red sandstone and conglomerate eroded off the Alps. The red sandstones of the Catskills in New York are similar rocks eroded off the ancient Appalachians. This suite of rocks is called molasse.

Ancient Shield Areas

Even though plate tectonics was accepted generally by about 1970, it took much longer for geologists to realize its full extent in time. Partly that was due to resistance by conservative opponents of plate tectonics, partly it was due to difficulty reconciling older concepts of crustal deformation like geosynclines with plate tectonics, and partly it was due to difficulty in recognizing the relationship between modern plate tectonic structures and their very deeply buried equivalents in ancient rocks. Geologists now generally accept that plate tectonics in very much its present form has operated for at least 2 billion years. However, there are also distinctive rocks found in the Precambrian that are rare or absent in later settings, and it appears that before 2.5 billion years ago, crustal processes might have been quite different from the present.

Thick, very pure quartzites are widespread in the mid-Precambrian. Celebrated examples include the Sioux and Baraboo Quartzites of North America and the Roraima Quartzites of South America (Angel falls, the highest waterfall on earth, drops off a high mesa of Roraima Quartzite). The mystery is how to accumulate huge thicknesses of very pure quartz sand. If the sand accumulated in a deep basin, we would expect high topography to result in rapid erosion and large amounts of other minerals. It is widely suspected that a carbon dioxide rich atmosphere created very acidic weathering conditions that rapidly destroyed other minerals.
Red sandstones and conglomerates ("red beds") are rare in the geologic record before 2 billion years ago. This is one of several indications that the atmosphere became more oxidizing about that time.
The most significant indicator of changed oxidation conditions around 2 billion years ago is the end of bedded iron formations. Bedded iron formations are common in early Precambrian rocks, interbedded with chert and deposited in marine sediments. Many of them appear to have been biogenically precipitated. The problem is that iron is extremely insoluble in today's oceans. To have been deposited in the quantities observed (pretty much the whole basis of industrial civilization) the iron would have to have been present as soluble ferrous iron (+2) which requires far less oxygen than we have at present. One theory is that iron accumulated in extremely anoxic deep ocean water, then brought to higher levels by upwelling where it was either metabolized by microorganisms or precipitated directly.
Uranium likewise has oxidation states that differ in solubility, except that oxidized uranium is soluble but reduced uranium is not, the opposite of iron. In present day conditions, uranium minerals are rapidly destroyed by weathering but in the Precambrian we find fossil placer deposits of uranium minerals, sometimes with pyrite sand grains. In at least one place (Oklo in the African nation of Gabon) a placer deposit accumulated a critical mass and underwent a natural chain reaction. The presence of detrital uranium, (not to mention pyrite!) is yet another indication that the early Precambrian atmosphere was oxygen poor.
Anorthosites are technically gabbro, in that they lack quartz and contain only plagioclase feldspar, but they consist almost entirely of feldspar. They make up much of the crust of the Moon, the result of gravitational differentiation during the Moon's magma ocean phase. Anorthosites occur in Archean rocks and are also widespread about 1.5 b.y. ago. Why, we don't know. It's tempting to consider these remnants of a magma ocean on the early earth, but they're always intrusive, so that hypothesis doesn't stand up.
Only small patches of continental crust older than 3 b.y. exist, but by about 2.5 b.y. ago, at least half of our present continental crust had formed. Most present day continents formed by accretion of terranes around ancient nuclei. The terrane accretion is a fairly straightforward analogue of more recent accretion, but the ancient nuclei have structures that are very hard to interpret in plate tectonic terms.
Large areas of the Archean cores of the continents (older than 2.5 b.y.) are made of granite-greenstone belt terranes. These consist of thick synclinal accumulations of volcanic and sedimentary rocks, intruded by granitic plutons. One a geologic map, they look like an old rag riddled with holes. Puzzling features of these terranes include:
- We never see any basement to the volcanic-sedimentary troughs. We never see what these rocks are resting on. The granite within and around them is always intrusive. The volcanic rocks are mostly basalt with abundant pillow lavas.
- The sedimentary rocks are usually greywackes deposited in deep water. There are some iron formations but no shallow water or terrestrial rocks. The absence of terrestrial rocks and granitic detritus has led to a suggestion that there was little topography during the Archean.
- Possible plate tectonic analogues of granite-greenstone belt terrains have been suggested. Some areas like the Sierra Nevada foothills consist of volcanic terrains intruded by granite plutons, and resemble granite-greenstone belt terrains. It has been suggested that continental nuclei with abundant granite-greenstone belt terrains formed by accretion of smaller individual terranes.
- On the other hand, some geologists have proposed that crustal processes during the Archean were fundamentally different, and that plate tectonics, if it happened at all, operated on smaller spatial and time scales.
Large areas of the Archean crust are made of gneiss, which indirectly suggest the existence of water and erosion, since they appear, on the basis of their composition, to have formed from sedimentary rocks.
Deep erosion of the continents (over 20 km in places) has almost entirely obliterated any traces of surface processes that could help us interpret the early earth. Deep crustal rocks like granulite have been exposed in places, especially where uplift has occurred.
Komatiites are ultramafic lavas, made mostly of olivine and pyroxene, with a distinctive dendritic texture (spinifex) under the microscope. They are mostly Archean, and the melting points of such lavas are well above any observed in more recent lavas. Komatiites are widely interpreted to mean the early earth was hotter at shallower depths than the present earth.

Return to Mineralogy-Petrology Index
Return to Thin-Section Index
Return to Crystals and Light Index
Return to Crystal Structures Index
Return to Mineral Identification Tables
Return to Professor Dutch's Home Page

Created 31 August 2011, Last Update