Tjeerd H. Van Andel

Morphology and tectonics of the inner rift valley at lat 36°50′N on the Mid-Atlantic Ridge

A segment of the inner rift valley of the Mid-Atlantic Ridge was investigated in detail from the American submersible Alvin . Fifteen traverses were made across the floor and up the first major fault scarps in the valley walls. The asymmetric morphology of the inner floor is found to be the primary result of volcanic activity modified by tectonic activity. Analysis of the tectonic features revealed that the rift is evolving within a single stress field that has its least principal strain axis (the compressional axis) aligned with the valley axis of N20°E. This is in contrast to the direction normal to plate divergence (N0°E). The tectonic elements in the inner floor are primarily vertically dipping tension fractures, whereas the fault scarps of the flanking walls are closer to a 60° dip and reflect a component of downdip shear. The information base obtained from Alvin was broadened with information collected in the area with more conventional techniques.nnThrough an analysis of this information, primarily the topography, it was possible to extrapolate the detailed observations obtained from the submersible to intervening areas to produce a comprehensive geological interpretation of the study area. An evolutionary model was developed which suggests that the inner rift is a product of axial volcanic activity. Shortly after formation, the original volcanic edifice is modified by vertical collapse, which leads to a reduction of the bottom relief. This process is reversed in the outer portions of the valley as uplift begins. Tensional extension changes into vertical shear as the volcanic blocks are incorporated into the walls and elevated. During the various stages of uplift, readjustment takes place on the terraces, which results in the preservation of the original volcanoes as recognizable units. This model, which spans 180,000 yr of inferred time, is examined in detail in an attempt to identify its weaknesses as well as to delineate the specific factual constraints upon which it is built. Alternate interpretations are proposed and tested in a similar fashion; the result is the identification of key problems that need to be solved.

Tectonics of the Panama Basin, Eastern Equatorial Pacific

The Panama Basin includes portions of the Nazca, Cocos and South America lithospheric plates and borders the Caribbean plate. The complex interactions of these units have largely determined the topography, pattern of faulting, sediment distribution, and magnetic character of the basin. Only heat flow data fail to correlate with major structural features related to these units. The topographic basin appears to have been created by rifting of an ancestral Carnegie Ridge. The occurrence of a distinctive smooth acoustic basement and a characteristic overlying evenly stratified sedimentary sequence on virtually all elevated blocks in the basin suggest that they all once formed part of this ancestral ridge. The present Carnegie Ridge is the relatively undeformed southern half of this feature, while the Cocos Ridge is the northern half fragmented by left-lateral north-south transcurrent faulting. As blocks of the Cocos Ridge reach the Middle America Trench, they appear to clog the subduction zone and become welded to the Nazca plate. Thus, the active transform fault at the eastern edge of the Cocos plate has episodically shifted west as segments of the trench were deactivated. Such a shift appears to be occurring at the present time.

Ponded Sediments of the Mid-Atlantic Ridge between 22° and 23° North Latitude

The numerous valleys of the flanks of the Mid-Atlantic Ridge are partly filled with sediment. The petrology of sediments from two such valleys (ponds) near 22° N. latitude shows that the material has been deposited by turbidity currents. The material has been derived from calcareous pelagic deposits which mantle the surrounding hills. The turbidite sequences show peculiarities which can be explained by assuming that the currents rebounded repeatedly from the pond walls. A theoretical model for the flow of these currents agrees well with density, thickness, and composition parameters that can be derived from the deposits and points to the occurrence of a hydraulic jump at the base of the slope. Adjacent valleys are separated by divides, and each valley contains an independent sedimentation unit. Good correlation exists between the volume of sediment in each valley and the size of the surrounding hill area from which the sediment was derived. Removal of 15 m of sediment from the entire region is required to account for the sediment that now fills the valleys. The oldest sediment outcropping in this region is upper Miocene, and the average sedimentation rate of the pelagic deposits can be estimated as approximately 3 mm/1000 years. In the ponds, the rate is much greater; more than 9 m of late Quaternary sediment have been found. Recently, the valleys have been faulted and partly uplifted. The accompanying earthquake activity may have initiated the turbidity currents responsible for the present rapid deposition on the valley floors.

The Vema fracture zone and the tectonics of transverse shear zones in oceanic crustal plates

At 11°N latitude, the Mid-Atlantic ridge is offset 300 km by the Vema fracture zone. Between the ridge offset, the fracture consists of an elongate, parallelogram-shaped trough bordered on the north and south by narrow, high walls. The W-E trending valley floor is segmented by basement ridges and troughs which trend W10°N and are deeply buried by sediment. Uniform high heat flow characterizes the valley area. Seismically inactive valleys south of the Vema fracture, also trending W10°N, are interpreted as relict fracture zones. We explain the high heat flow and the shape of the Vema fracture as the results of secondary sea-floor spreading produced by a reorientation of the direction of sea-floor spreading from W10°N to west-east. This reorientation probably began approximately 10 million years ago. Rapid filling of the fracture valley by turbidites from the Demerara Abyssal plain took place during the last million years.The large amount of differential uplift in the Vema fracture is not explained by the reorientation model. Since the spreading rate across the valley is small compared to that across the ridge crest, we suggest that it takes place by intrusion of very thin dikes that cool rapidly and hence have high viscosity. Upwelling in the fracture valley will thus result in cosiderable loss of hydraulic head, according to models by Sleep and Biehler (1970), and recovery of the lost head could produce valley walls higher than the adjacent ridge crest. We further postulate that the spreading takes place along the edges of the fracture zone rather than in the center. This would account for the uniform distribution of heat flow along the fracture valley and for the lack of disturbance of the valley fill. As a consequence, a median ridge should form in the center, where head loss is compensated in the older crust; such a median ridge may be present. The width of the valley should be a function of the angle and time of reorientation, and of the spreading rate; the width so obtained for the Vema fracture is in accordance with the observed width. If this model is correct, the narrowness of the valley walls implies a thin lithosphere of very limited horizontal strength.

Rifting origin for the vema fracture in the North Atlantic

The mid-Atlantic ridge crest is offset more than 300 km along the E-W trending Vema Fracture valley. The trend of the extremeties of this valley outside the region of recent earthquakes, as well as that of apparently inactive transverse valleys to the south, is somewhat south of east. These observations, and the fact that the valleys to the south apparently do not extend across the ridge crest, strongly suggest that a new pattern of ridge growth and sea-floor spreading, represented by the Vema Fracture valley, has been superimposed on an older one. A geometrical model is proposed for the recent development of the Vema Fracture valley, which is shown to be consistent with other geological and geophysical observations in this region.

Cenozoic Migration of the Pacific Plate, Northward Shift of the Axis of Deposition, and Paleobathymetry of the Central Equatorial Pacific

Cenozoic northward migration of the Pacific plate is documented by magnetization vectors of seamounts and by volcanic lineations resulting from drift over fixed melting spots in the mantle. The rotation with respect to the spin axis of the Earth can also be established from the northward shift with time of the equatorial axis of maximum deposition. Data from Deep Sea Drilling Project sites in the equatorial Pacific indicate a pole of rotation for the past 45 m.y. at lat 67° N., long 59° W., in satisfactory agreement with locations derived from other evidence. The best fit is obtained for an initial rate of 0.25°/m.y., which accelerated to 0.8°/m.y. about 25 m.y. B.P. With this rotation scheme and the subsidence history of the individual drill sites, the paleobathymetric evolution of the central equatorial Pacific during Cenozoic time and the position of the ancestral East Pacific Rise can be established.

Cenozoic Calcium Carbonate Distribution and Calcite Compensation Depth in the Central Equatorial Pacific Ocean

Using an absolute time scale for the stratigraphic sections penetrated in 20 drill sites of the Deep Sea Drilling Project in the eastern equatorial Pacific, we have determined paleodepths, paleopositions, and average CaCO 3 content for each 1-m.y. interval for the past 50 m.y. The calcite compensation level (CCD) at the equator was very shallow (3,000 m) during Eocene time, dropped precipitously to a low at 5,000 m in Oligocene time, then rose to its present position. The CCD outside the equatorial zone was also shallow during Eocene time, but while the equatorial CCD remained depressed during the Oligocene, the extra-equatorial CCD gradually shoaled. This suggests an increase in carbonate dissolution in the Pacific, matched by an increase in carbonate supply at the equator. During Miocene time, both the equatorial CCD and the Pacific CCD rose together. Finally, near the Miocene-Pliocene boundary, the extra-equatorial CCD dropped from its Miocene high of at least 3,900 m to its present level.

Basement Ages and Basement Depths in the Eastern Equatorial Pacific from Deep Sea Drilling Project Legs 5, 8, 9, and 16

Recent literature contains numerous references to basement ages and basement depths determined by the Deep Sea Drilling Project. The data are derived from a variety of sources, many of them inadequately documented or preliminary, and are not uncommonly inaccurate or conflicting. In this paper we present tabulations of basement ages and depths from DSDP Legs 5, 8, 9, and 16 in the eastern equatorial Pacific, refer them to the latest biostratigraphic time scale, and document and discuss their error limits. We recommend that in future use of this type of data a similar practice be adopted and that the precise source of data, time scale used, and procedures for determining ages and errors be clearly identified in order to avoid confusion. Based on the data presented here, we also give the relations between basement age, distance from the spreading center, and basement depth. The errors inherent in the data cause these relations to be very general and to have less resolution than ascribed to them by some previous investigators.

Ascension Fracture Zone, Ascension Island, and the Mid-Atlantic Ridge

Near 7° S. latitude, the Ascension fracture zone offsets the Mid-Atlantic Ridge right-laterally over 230 km. North of the fracture zone, which trends about N. 80° E., the ridge crest is perpendicular to its trend, but to the south, near 8° S., the initially perpendicular trend changes to nearly northerly. Ascension Island lies approximately 50 km south of the fracture on magnetic anomaly 4, with an inferred age of 7 m.y. It is not on any major tectonic trend and there is no evidence that it is part of a volcanic chain. Spreading rates in the region increase from north to south, proportional to the distance from the pole of rotation of the African and South American plates, and may be slightly different on the east and west sides of the ridge. Normal to subnormal heat-flow values prevail except for one high value east of the northern ridge axis. The Ascension fracture valley is wide and filled with thick sediments implying an anomalously high age. Earthquake epicenters are aligned along the ridge crest, but near the fracture zone they define an activity belt south of it and more nearly east-west trending. The data suggest a shift of the fracture zone to an east-west trend about 10 m.y. ago, followed by a reorientation of the southern ridge axis that proceeded from south to north and has not been completed. The hypothesis accounts for most observations except the heat-flow pattern, the absence of epicenters on the southernmost ridge crest, and some small structural features.

Tectonics of the Mid-Atlantic Ridge, 6–8° South latitude

A regional geophysical traverse of the Mid-Atlantic Ridge in the northern South Atlantic was obtained during CIRCE cruise of the Scripps Institution of Oceanography. During the traverse, four detailed surveys were made of small areas on the crest and east flank. The geomagnetic anomaly profile can be used as a time base for the interpretation of tectonic events of the ridge. The profile also suggests that the rate of sea-floor spreading in this part of the South Atlantic accelerated twice, approximately 40 and 4.5 million years ago, and decelerated at least twice, 38 and 10 million years ago. Accelerations were probably accompanied by uplift and normal faulting of the central part of the ridge, while decelerations produced subsidence with modest contraction, reflected in reverse faulting and folding. The effects of uplift are clearly present in the reflection seismic records, which are, however, not well suited to detect reverse faulting.Spreading without creation of significant relief occurred on the ridge until approximately 5 million years ago. This process produced a low relief with small rifts, strongly reminiscent of the present crestal topography of the East Pacific Rise. A markedly linear secondary relief of 100–200 m, parallel to the ridge axis, developed later by faulting of the flanks. Portions of the crust that were near the crest during periods of uplift are more intensely faulted than those that were remote at all times. The importance of the last uplift of the crest and associated faulting on the flanks is reflected by a decrease in the density of faulting away from the ridge crest.The present crestal zone is very different from the flanks and from the older crests; the relief is nearly ten times greater, transverse disturbances are common, and there is conflicting evidence regarding its age. This striking change in character indicates either a recent change in the spreading process or a recent period of strong deformation which has affected only the crestal zone.