Walter C. Pitman

Plate Tectonics and the Evolution of the Alpine System

It is contended that the Late Triassic to present-day gross evolution of the Alpine system in the Mediterranean region has been the result of activity along an evolving network of accreting, transform, and subducting plate boundaries between the large stable cratons of Europe and Africa. A refined assembly of the outlines of the continents around the North and central Atlantic, before the initial dispersion of Gondwanaland in Early Jurassic times, is presented. By considering geologic facies, structural fabric, and paleomagnetic criteria, the smaller continental fragments now found within the Alpine system are restored to their proposed initial positions relative to each other in the reconstruction offered. The motion of the major plate of Africa relative to Europe, commencing with the initial continental fragmentation, is documented by analysis of the sea-floor spreading history of the Atlantic Ocean, with the assumption that plate accretion there has occurred between torsionally rigid lithospheric plates. By the computerized fitting of well-defined and well-dated key pairs of symmetric magnetic anomaly lineations back together by a series of finite rotations, the relative position of North America to both Europe and Africa has been determined for the following times: 180 m.y. (Toarcian Stage, Early Jurassic); 148 m.y. (Kimmeridgian Stage, Late Jurassic); 80 m.y. (Santonian Stage, Late Cretaceous); 63 m.y. (Danian Stage, Paleocene); 53 m.y. (Ypresian Stage, Eocene); and 9 m.y. (Tortonian Stage, Miocene). From these positions, a series of rotation poles presumed to describe the stepwise motion of Africa relative to Europe were computed. The motions of the smaller intervening microplates have been inferred from the style of tectonic deformation on their borders, and these motions have been constrained to satisfy both changes in paleo-latitude with time and progressive rotations relative to the large macroplates that can be deduced from paleomagnetic measurements. The evolution of Tethys does not involve a single simple plate boundary between Europe and Africa, as has been envisioned previously, but, instead, a constantly evolving mosaic of subsiding continental margins, migrating mid-oceanic ridges, transform faults, trenches, island arcs, and marginal seas (back-arc basins). The periods of passive-continental margin development are recognized by a transgressive facies of platform carbonate rocks and thick prisms of continental-rise type sedimentation; accreting ridges by ultramafic rocks, gabbro, pillow basalt, deep-sea pelagic ooze, and abyssal red clay of the ophiolite suite; trenches by a migrating series of progressively younger linear flysch troughs whose immature mineral composition reflects nearby andesitic and metamorphic source terrains; the arcs themselves by calc-alkaline volcanism and the intrusion of silicic to intermediate plutons; the polarities of these arcs by the direction of overthrust nappe sheets and gradients in the ratio of potash to silica in the extrusives; their orientation by paired belts of high T and P and high P-T metamorphics; and finally the spreading back-arc basins by outpourings of basaltic magmas and evidence of flipping Benioff planes. A compilation of eight phases or chapters in Atlantic spreading history are outlined, which are based on the recognition of discrete differences and (or) relative motion between the continents bordering the Atlantic. All of these changes are reflected in the Tethys by reorganizations of the intervening plate boundaries and, we believe, are most explicitly recorded in the deformational history of the subducting zones. A montage of geometrically assembled plate-boundary interpretations are pictorially displayed as time-lapse frames of the evolving Alpine system. The montage begins with the Late Triassic (pre-Atlantic) setting of the Tethys 1 Ocean and extends to the present through nine phases of Tethyan history. Each phase is recognized on the basis of the age of intrusion and extrusion of basic lavas in ophiolite complexes, which mark the creation of new oceanic areas by both axial accretion in rift valleys of mid-oceanic ridges between rigid plates or by a more uncertain type of spreading in basins behind active island arcs. All the schemes presented are best estimates of the gross geometrical arrangements at discrete time intervals and should be treated as merely educated guesses. Despite the fact that we only have rigorous constraints for the relative positions of the nondeformed forelands of Europe and Africa, our models nevertheless imply that the motions of the larger plates will, by and large, dictate the general behavior of the smaller microplates through the particular styles of deformation set up along the adjoining plate boundaries. The Tethys 1 Ocean, located between Africa and Europe in Triassic times, has been almost entirely swallowed up in subduction zones of the Major Caucasus Mountains along its former northern margin and in similar zones of the Pontides and Minor Caucasus along its southern margin. The only remnants of Tethys 1 are the areas of oceanic crust in the Black and South Caspian Seas. There is considerable evidence to suggest that the Tethys 1 Ocean had an actively spreading ridge. Some tens of millions of years prior to the opening of the central North Atlantic, a branch of this ridge system entered into the Vardar Zone of eastern Greece and broke off fragments of northeast North Africa to initiate the development of the present-day Ionian and Levantine Basins of the eastern Mediterranean. Additional fragments (the Moroccan and Oranaise Meseta) were ruptured from northwest Africa following its separation from North America. The intervening Jurassic Atlas, seaway developed along an accreting plate boundary extending from the eastern Tethys to the crest of the embryonic Mid-Atlantic Ridge where it formed a migrating triple junction whose trace, we believe, follows the trend of the New England seamount chain. The western Mediterranean basins of the Alboran, Balearic, and Tyrrhenian Seas are very much younger, being initially opened in the early Miocene as a string of back-arc marginal seas behind the developing Apennine, Tel Atlas, and Rif suture zone that today marks the sites of subduction of Jurassic and Lower Cretaceous oceanic crust. The contemporary Alpine system displays a spectrum of stages in the building of mountain belts. Embryonic nappes within the Mediterranean Ridge in proximity to melange zones of the inner wall of the Hellenic Trench are, perhaps, signs of the initial deformation of sedimentary passengers on oceanic crust arriving at a subduction zone. Total closure of an ocean followed by the partial consumption of a passive continental margin leads to events such as the tectonic emplacement of crystalline basement nappes of the European “chaine calcaire” onto northwest Africa. Arc-continent collisions of this type which have then been succeeded by total destruction of marginal back-arc basins are recognizable in the Hellenides and Pontides. There are, as well, collisions that have not involved the disappearance of large oceanic areas; these are most apparent in the particular tectonic style of the Pyrenees and High Atlas Mountains.

Sea-Floor Spreading in the North Atlantic

The magnetic anomaly lineation pattern in the North Atlantic Ocean (between the latitudes of 15° N. and 63° N.) has been examined in light of the hypotheses of sea-floor spreading and plate tectonics. There is no evidence of significant subduction or deformation along the margins of the Atlantic since the Late Triassic, and thus the sea-floor spreading that has occurred since that time has resulted in continental drift only. The rate and direction of drift between Europe and North America and between Africa and North America have differed at all times since the Late Triassic. Although Eurasia may have been rifted from North America in the Jurassic, the major phase of drift did not begin until the Late Cretaceous. Separation varied from 5.0 to 4.0 cm/yr (at a latitude of 45° N.) from the Cretaceous until 53 m.y. ago. The rate of separation slowed about 53 m.y. ago. The average rate was slightly less than 2 cm/yr for the intervals from 53 m.y. to 38 m.y. ago and from 38 m.y. to 9 m.y. ago. The sediment discontinuity found by others at about the location of anomaly 5 on both flanks of the Mid-Atlanti.c Ridge, north of the Azores, thus cannot be explained by a discontinuity or drastic slowing in the rate of spreading. From 9 m.y. to the present, separation has been at a rate somewhat greater than 2.0 cm/yr. The initiation of rifting between Africa and North America may have occurred 200 m.y. ago. However, we have assumed that the active phase of drift did not begin until 180 m.y. ago. The separation proceeded at an average rate of 4.0 cm/yr from 180 m.y. to 81 m.y. ago; 3.4 cm/yr from 81 m.y. to 63 m.y. ago; 2.4 cm/yr from 63 m.y. to 39 m.y. ago; 2.0 cm/yr from 38 m.y. to 9 m.y. ago; and 2.8 cm/yr from 9 m.y. ago to the present (the rates are computed for a latitude of 35° N.). We have fitted together lineations of the same age but from opposite sides of the ridge axis in the same fashion that previous workers have fitted together continental margins. Each fit is described by a pole and angle of rotation about the pole. Each fit gives the paleogeographic relations of the respective continents and oceanic plates for the particular age of the lineation. We conclude from these paleogeographic reconstructions that there was probably no Late Cretaceous (81 m.y. to 63 m.y. ago) sea-floor spreading in the Arctic, but that the relative motion between Eurasia and North America in the Arctic region was compressional during this interval. This compression may have been accommodated by subduction at Bowers Ridge (which appears to be an inactive island-arc trench system) and subduction in eastern Siberia. It also may have been accommodated by compressional deformation in the Brooks Range, the Verkhoyansk Mountains, and the Sverdrup Basin (in central northern Canada). All the spreading in the Arctic region that has occurred since the Late Cretaceous has taken place in the last 63 m.y. The locus of this spreading has been the Mid-Arctic Ridge which lies between the Lomonosov Ridge and the Eurasian continental shelf. The effect of this spreading has been to separate the pre-existing Lomonosov Ridge from the Eurasian continental shelf. The Alpha Cordillera has not been the locus of sea-floor spreading in the Cenozoic. The exact pattern of the separation of Greenland from North America is not known. There may have been minor rifting in the Labrador Sea during the Jurassic. However, the major phase of drift occurred from the Late Cretaceous to the late Eocene. The final separation of Eurasia (Spitsbergen), Greenland, and North America did not occur until the middle Eocene. The pattern of magnetic lineations suggests that the well-documented counterclockwise rotation of the Iberian Peninsula occurred between the Late Triassic and the Late Cretaceous, and that there has been little, if any, counterclockwise rotation subsequent to that time. We have used the derived poles and the angular rates of rotation to compute isochrons which give the age of the basement in the North Atlantic. The basement ages agree well with other data such as those obtained as the result of JOIDES drilling. The isochrons sometimes give greater ages which can be reconciled with the drilling results by involving subsequent volcanism, but in no case do the isochrons give smaller ages. The Keathley sequence of magnetic anomalies which lie just seaward of the quiet zone and southwest of Bermuda in the western Atlantic and northwest of Dakar in the eastern Atlantic, has been given an age of about 130 to 155 m.y. Comparison of the isochrons with the magnetic lineations indicate that two important shifts of the ridge axis may have occurred. The first, in the region south of the New England Seamounts and the Canary Islands was a 200-km eastward jump or migration that took place prior to 155 m.y. ago; the second in the region north of the New England Seamounts and Canary Islands but south of the Azores was a more complex westward shift of 150 km maximum extent that occurred between 135(?) m.y. and 72 m.y. ago. We have also computed a pattern of synthetic fracture zones or flow lines. Previous workers have proposed that the South Atlas fault, the western Canary Islands, and the New England Seamounts lie along a fundamental fault or fracture zone. We note that these features are approximately parallel to one of these synthetic flow lines. The seaward escarpment bounding the southern Bahamas as well as several well-surveyed fracture zones and other bathymetric features are parallel to the synthetic fracture zones.

Relationship between eustacy and stratigraphic sequences of passive margins

It is commonly thought that transgressive or regressive events that may have occurred simultaneously on geographically dispersed continental margins have been caused by worldwide sea-level rise or fall, respectively. Instead, it is shown here that these events may be caused by changes in the rates of sea-level rise or fall. The subsidence of an Atlantic-type (passive) margin may be modeled as a bordering platform rotating downward about a landward hinge line. The rate of subsidence is greatest at the seaward side of the platform and decreases landward to zero at the hinge line. With the exception of sea-level changes due to glaciation, dessication, and flooding of small ocean basins and other sudden events, the rate of subsidence at the seaward edge of the platform (shelf edge) is greater than the rate at which sea level may possibly rise or fall. Thus, if sea level is falling, the shoreline will seek that point on the subsiding platform at which the rate of sea-level fall is equal to the rate of subsidence minus the sedimentation rate. If the rate of sea-level fall decreases, the shoreline will move landward; if the rate increases, the shoreline will migrate seaward. If sea level is rising, the shoreline will move to that point where the rate of sea-level rise is equal to the sedimentation rate minus the subsidence rate. Thus, if the rate of sea-level rise decreases, the shoreline will move seaward; if the rate increases, the shoreline will move landward. The position of the shoreline is also a function of the sedimentation rate. These relationships have been quantified so that the position of the shoreline and the thickness of the sediments deposited during discrete time intervals may be computed as a function of the rate of sea-level change and the sedimentation rate. A sea-level curve, based on volume changes of the mid-oceanic ridge system, has been computed. Sea level is seen to fall persistently from Late Cretaceous to middle Miocene time, but transgressions occur in Eocene and early Miocene time because the rate of sea-level fall is slower for these periods. It is concluded also that the presence of the shoreline seaward of the shelf edge of an Atlantic margin should be symptomatic of events that may cause rapid sea-level fall, such as glacial build-up or the sudden flooding of large deep basins.

World-Wide Correlation of Mesozoic Magnetic Anomalies, and Its Implications

In the course of correlating three sets of Mesozoic magnetic lineations in the western Pacific (the Phoenix, Japanese, and Hawaiian lineations), Larson and Chase (1972) determined a paleomagnetic pole for the Pacific plate for the Early Cretaceous. Using this pole we have derived a magnetic reversal model for the Hawaiian lineation set. We then have used this model to correlate the entire Hawaiian lineation set to the entire Keathley lineation set in the western North Atlantic. On the basis of these correlations and drill holes associated with the lineation patterns, we have extended the geomagnetic reversal time scale back to the base of the Late Jurassic (162 m.y. B.P.). A period of reversals occurred corresponding to the Hawaiian and Keathley lineations from 150 o t 110 m.y. B.P., and these reversals are bracketed by long periods of dominantly normal polarity (the Cretaceous and Jurassic magnetic quiet zones). This magnetic reversal time scale significantly alters previous notions of the timing and origin of sea-floor spreading features in the Atlantic Ocean. It implies that the Bay of Biscay opened sometime during the interval between 150 and 110 m.y. B.P.; that drift in the South Atlantic was initiated at sometime during the interval from 110 to 85 m.y. B.P. (probably close to 110 m.y. B.P.); and that the seaward portion of the marginal quiet zones of the eastern United States and northwestern Africa resulted from sea-floor spreading during the Late Jurassic period of dominantly normal magnetic polarity prior to 150 m.y. B.P. In the Pacific during the late Mesozoic, spreading was occurring from at least five spreading centers joined at two triple points. The vast majority of the Pacific Basin today is occupied by only the Pacific-plate side of these spreading patterns. This implies that an area equal to most of the Pacific Basin has been subducted beneath the surrounding continents since the Early Cretaceous. Our magnetic reversal time scale calls for a rapid pulse of spreading from about 110 to 85 m.y. B.P. at all the spreading centers in both the Atlantic and Pacific Oceans. This implies a pulse of rapid subduction around the rim of the Pacific that we relate to episodes of large-scale plutonism in eastern Asia, western Antarctica, New Zealand, the southern Andes, and western North America during the Late Cretaceous.

A plate-kinematic framework for models of Caribbean evolution

Abstract We define the former relative positions and motions of the plates whose motions have controlled the geological evolution of the Caribbean region. Newly determined poles of rotation defining the approximate spreading histories of the central North and the South Atlantic oceans are given. For the late Jurassic-Early Cretaceous anomaly sequence of the central North Atlantic, we have used previously published ∗ definitions of fracture-zone traces and magnetic anomaly picks, redetermining the pole positions and angular rotations for various isochrons on an Evans and Sutherland interactive graphics system. For magnetic anomalies younger than the Cretaceous Quiet Period in both oceans, we (1) used Seasat altimeter data to help define fracture-zone traces, and (2) identified and used marine magnetic anomalies to determine the positions of spreading isochrons along the flowlines indicated by the fracture zones. By the finite difference method, the relative paleopositions and the relative motion history between North and South America were computed. This analysis defines the size and shape (and the rate at which the size and shape changed) of the interplate region between North and South America since the Middle Jurassic. Thus, a plate-kinematic framework is provided for the larger plates pertaining to the Caribbean region, in which can be derived more detailed scenarios for Gulf of Mexico and Caribbean evolution. North and South America diverged to approximately their present relative positions from Late Triassic? to Early Campanian (about 84 m.y. ago) time. This is the period during which the Gulf of Mexico and a Proto-Caribbean seaway were formed. Since the Campanian, only minor relative motion has occurred; from Early Campanian through to Middle Eocene times. South America diverged only another 200 km, and since the Middle Eocene, minor N-S convergence has occurred. These very minor post-Early Campanian motions have probably been accommodated by imperfect shear and compression along the Atlantic fracture zones to the east of the Lesser Antilles, and along the northern and southern borders of the Caribbean Plate. Accordingly, it is suggested that from Campanian time to the present, the relative motions between the North and South American plates have had only minor effects on the structural development of the Caribbean region. Primarily using the data of Engebretson et al. ∗∗ , the convergence history of Pacific plates with North America was calculated for two points near the western Caribbean. By completing finite difference solutions, the convergence history of the Pacific plates with the Caribbean and South American plates can be approximated. The direction and rate of convergence of the Pacific plates with the Americas may have controlled the style of subduction and possible microplate migration along the North American, South American and western Caribbean boundaries that define the eastern Pacific plate margin.

Plate Tectonics Model for the Evolution of the Arctic

Following a search of critical literature on the geology and geophysics of the Arctic, we have constructed a model for the post-Permian evolution of the Arctic Ocean that follows the tenets of plate tectonics. We consider the history of the Arctic as the study of two separate basins, the Cenozoic Eurasian Basin and the Mesozoic-Cenozoic Amerasian Basin, and we have utilized the detailed pattern of opening of the North Atlantic worked out by Pitman and Talwani (1972) to determine the relative positions of Eurasia and North America during the past 81 m.y. We propose that the Amerasian Basin as we now know it opened by sea-floor spreading during the Jurassic magnetic quiet period, 180 to 150 m.y. ago. We reject the interpretation of the Alpha-Mendeleyev Ridge complex as an early Cenozoic spreading center and show that this feature is better interpreted as a fossil subduction zone-incipient island arc. The Eurasian Basin is an extension of the North Atlantic, which has opened by sea-floor spreading during the past 63 m.y. Prior to 63 m.y., the Lomonosov Ridge formed the seaward edge of the Eurasian continental margin.

Age of the North Atlantic Ocean from magnetic anomalies

Abstract Magnetic anomaly lineations have been identified in the North Atlantic. These lineations correlate with the magnetic time scale describing magnetic polarity reversals for the past 71 my. The results indicate that approximately 70% of the total drift between Europe and North America has occurred in the past 72 my, whereas only about 35% of the total drift between Africa and North America has happened in the same period. Extrapolation using the magnetic anomaly data and the results of JOIDES drilling suggests that drift between Africa and North America was initiated about 180 mybp.

The Age of the South Atlantic

This paper attempts to further delineate the pattern of continental drift that has taken place in the South Atlantic since mid-Mesozoic time. The history presented is based on the analysis of the magnetic anomaly pattern associated with the mid-Atlantic ridge which indicates the age of the basement rock and may be used to determine to a first approximation the pattern of drift between the continental masses that surround the Atlantic.

Thermomechanical Properties and Evolution of Small Pull-Apart Basins: ABSTRACT

Small pull-apart basins are generally characterized by 2 component subsidence: an initial essentially instantaneous isostatic subsidence (Si) dependent on the ratio of crustal to lithospheric thickness (Cz/lz) and the stretching factor s, followed by a slower decaying thermal subsidence (St) controlled by the thermoelastic properties of the continental lithosphere, which in turn can be characterized by a thermal time constant ^tgr. Rapid short-lived subsidence (e.g., Vienna basin, Californian Miocene basins) is indicative of either (1) inhomogeneous crustal stretching without major sublithospheric involvement, or (2) extremely small lithospheric diffusivities. The former implies a thin-skinned origin for pull-apart basins nd suggests that the spatial and temporal distribution of bounding faults and splays typical of pull-apart basins, result from inhomogeneous brittle failure of the upper crust. However, the effects of lateral heat flow decrease the thermal time constant by allowing a basin to subside more quickly due to both lateral and vertical cooling. The size of this effect is dependent on the width of the stretched lithosphere (the effective ^tgr of a 100 km wide rift is 36 m.y., for a 25 km rift, 6 m.y., whereas the actual thermal time constant in both cases is 62.8 m.y.). Lateral heat flow amplifies rift subsidence while producing complementary time-transgressive uplift in adjacent unstretched regions. However, the flexural rigidity of the lithosphere severely attenuates the deformation caused by he lateral flow of heat. Whereas the deformation is highly dependent on the mechanical properties of the lithosphere, ^tgr is independent. Continental lithospheric rigidities appear to increase with age following an orogenic or thermal event, suggesting that the long-term mechanical behavior of the continental lithosphere is similar to that of the oceanic lithosphere. However, high rigidities (1032 dyne-cm) associated with Archean/Proterozoic terranes and modeling of plate deformation suggest that the long-term thermal behavior of continental lithosphere is governed by a cooling plate model with a 200-250 km lithospheric thickness, nearly twice the 125 km estimates for the oldest oceanic lithosphere. This has important implications for the evolution of sedimentary basins. A doubling of the lithospheric thickness implies a quadrupling of ^tgr, yet basin subsidence models have assumed that ^tgr for the oceanic a d continental lithospheres are similar. A large ^tgr allows basin subsidence to continue over significantly longer times, but now lateral heat flow, in addition to vertical, must be included in basin models to obtain accurate subsidence and temperature estimates. In particular, Si is highly dependent on the age of the underlying basement. These principles are illustrated both theoretically and with reference to the European Alpine foreland, upper Paleozoic foreland basins of North America, and Californian Neogene basins. End_of_Article - Last_Page 794------------