GEOLOGY AND GEOPHYSICS OF MID-ATLANTIC RIDGE PHYSIOGRAPHIC PROVINCES
Seismicity of the Mid-Atlantic Ridge.—The earthquake epicenters instrumentally determined for the North Atlantic up to 1956 are shown in Plate 29. Nearly all earthquakes fall in the crest zone. Considering that determination of shocks is accurate only to within -½° to 1° +, it is quite surprising that the plotted epicenters form such a narrow belt. Investigation of the problem of the physiographic province most seismically active reveals that many epicenters actually plot in the Rift Valley and that virtually all that do not are within about 1° of the Rift Valley. All seismic activity therefore is limited to the crest provinces, and probably virtually all the activity is concentrated within the Rift-Valley Province. A line of epicenters runs from the Rift Valley near Flores Island of the Azores toward the Straits of Gibraltar.
Sediments and physiographic provinces of the Mid-Atlantic Ridge.—The Mid-Atlantic Ridge is the major site of undisturbed pelagic sedimentation of the Atlantic because of its isolation from down-slope movements starting on the continental margin. However, turbidity currents must form near the shores of oceanic islands and the edges of shallow banks and probably contribute sediments to the intermontane valleys. Photographs taken on the Rift Mountains and on the sides of major seamounts show scour marks and ripple marks, indicating considerable winnowing and scour by deep-ocean currents (Pl. 19). This sediment carried from the tops of peaks and deposited on the steep mountain sides probably slumps occasionally and forms turbidity currents which flow to the adjacent valley floors (Pl. 28). For this reason cores taken in intermontane valleys and near the higher peaks will have considerable interlayering of turbidity-current deposits, and much of the sides and crests of individual high mountains is bare rock. In the Rift Mountains true pelagic sediments are only occasionally found. It is striking to note that in coring and bottom photography bare-rock slopes are found most commonly in the Rift Mountains and High Fractured Plateau, but flat-floored intermontane basins are absent in these provinces. This must indicate either that the topography of the Rift Mountains is very new or that the sediment eroded from the crest provinces is carried all the way to the Upper Step Province where it is deposited in the intermontane basins.
Rocks of the Mid-Atlantic Ridge.—Our knowledge of the lithology of the Mid-Atlantic Ridge comes from three sources: (1) rocks dredged from the sea floor, (2) detrital rock fragments found in sediment cores, and (3) rocks exposed on the islands of the Ridge.
Some of the earliest rock dredging on the Mid-Atlantic Ridge was done in 1885 by the Talisman expedition. In 1949 Furon (1949) reported the occurrence of fossil trilobites in dredge samples which had been stored for more than half a century in a French Museum. One dredging was made in the High Fractured Plateau of the eastern Atlantic at 42° 21´N., 17° 12´W., in 4255 meters depth (2330 fathoms). Furon believes that the material was in situ and therefore proof of early Paleozoic outcrops on the Mid-Atlantic Ridge. The abundant evidence of glacially rafted rocks even as far south as 30° N. casts serious doubt on this conclusion, but nevertheless the possibility that the material might have been in situ must be considered.
The Mid-Atlantic Ridge Expedition of 1947 led by Ewing made a number of successful rock-dredge hauls on the Mid-Atlantic Ridge. The most successful hauls were made in the Rift Valley and on the adjacent Rift Mountains at about 30° N. Lat. The specimens have been described by Shand (1949), who reported olivine gabbro, serpentine, basalt, and diabase. One limestone of probably Tertiary age was collected the same year but has not been described. The suite of crystalline rocks obtained is similar to those found on oceanic islands elsewhere on the Mid-Oceanic Ridge.
Crustal structure and the Mid-Atlantic Ridge provinces.—The crustal structure of the Mid-Atlantic Ridge provinces has been determined at about 20 places by the seismic-refraction technique (Fig. 35f). These studies, conducted by John I. Ewing and W. M. Ewing (in press), have shown that the average crustal structure of the crest provinces and Upper Step consists of 0.4 km of low-velocity sediment and 2.8 km of rock with a velocity of 5.1 km/sec overlying a substratum in which the velocity is 7.3 km/sec. The thickness of the layer of low-velocity sediment varies considerably from place to place. In the crest provinces the 5.1 km/sec layer is commonly exposed. In the flank provinces appreciable thicknesses (to 1 km) of sediment have been measured. The sediment seems to thicken between major scarp zones and ridges as if the sediment were collecting in longitudinal basins parallel to the axis of the ridge. An insufficient number of measurements have been made to determine whether these accumulations correlate with the boundaries of individual intermontane basins or with the limits of individual step provinces.
The structure of the abyssal floor, crest provinces, and flank provinces is compared in Figure 35. The two higher-velocity layers shown in the abyssal-floor sections (Nares and Sohm Abyssal Plain, Fig. 35), have been observed in all measurements made in these provinces. The 6.7 km/sec layer is generally considered to be gabbroic, and the 8.1 km/sec layer is by definition the earth's mantle. In the Mid-Atlantic Ridge section the upper high-velocity material has an average velocity of 5.1 km/sec and is generally identified as basaltic rock. The velocity of the underlying material (7.3 km/sec) is intermediate between the velocity of oceanic crustal rocks (6.7 km/sec) and that of mantle rocks (8.1 km/sec), as observed both beneath the continents and beneath the abyssal floor.
Ewing and Ewing (in press) suggest that this intermediate velocity is the result of a physical mixture of oceanic crustal rocks and mantle rocks. To explain such large-scale mixing they propose that extensive vulcanism and intrusion along the Mid-Atlantic Ridge have produced an intermingling of the crustal and mantle rocks, and that this was associated with convection cells in the deep mantle which supply large quantities of basaltic magma and produce extensional forces on the crust and upper mantle.
Nearly 20 crossings of the crest of the ridge have been made with the total-intensity magnetometer towed behind research vessels employing continuously recording echo sounders. A characteristic anomaly pattern has been noted by Ewing, Heezen, and Hirshman (1957). The Rift Valley is characterized by a large positive anomaly, while the adjoining Rift Mountains show negative anomalies of 300 to 500 gammas (Fig. 48).
Free-air gravity anomalies over the crest provinces and Upper Step are usually 30-50 mg positive, while the Rift Valley as measured in two places gave free-air anomalies of -3 and -20 mg.
Figure 48.—Profile of total magnetic intensity and topography, Mid-Atlantic Ridge
Soundings made with PDR. Magnetic measurements made with fluxgate total-intensity magnetometer. Magnetic values in gammas relative to an arbitrary zero.
Figure 49.—Physiographic provinces and trans-Atlantic structure
Based on scattered seismic-refraction measurements in the North Atlantic which have been projected along province boundaries. The topographic profile was pieced together from continuously recorded echo-sounding profiles from New York to Spanish Sahara.
In the continental margin the upper layer represents the sedimentary rock. The dashed symbol indicates the continental crustal rocks. The lower layer represents oceanic crustal rocks. The mantle lies below the lowest layer.
A heat-flow measurement by E. C. Bullard in the Rift Valley province in the North Atlantic indicated a value of about 7 × 10-6 cal./cm2/sec. which is about 6 times the average value of 1.2 × 10-6 cal./cm2/sec observed in the Lower Step and abyssal floor of the eastern Atlantic (Bullard, 1954; Bullard et al., 1956).
High heat-flow values have also been observed on the Easter Island Ridge of the Southeast Pacific, suggesting that the entire Mid-Oceanic Ridge rift system may be so characterized.
An adequate synthesis and explanation of all these converging lines of evidence has not yet been formulated. However, the correlation of so many types of geophysical and geological data speaks favorably for the validity and tectonic significance of the physiographic provinces described here.
On the basis of the observed correspondence of crustal structure and physiographic provinces a hypothetical trans-Atlantic structure section was prepared (Fig. 49). Seismic-refraction measurements were projected along province boundaries and plotted beneath an echo-sounding profile from New York to Spanish Sahara. The black splotched areas represent the 7.3 km/sec layer. This velocity intermediate between 8.1 km/sec of normal mantle and 6.7 km/sec of normal oceanic crust is considered (1) a mixture of the two normal layers; (2) a low-velocity part of the mantle, or (3) a distinct crustal layer characteristic of mid-oceanic ridges. The structure shown for the continental margin of Africa is based on analogy with the structure of the continental margin of northeastern United States. This procedure seems justified by the close similarity of the continental-margin physiographic provinces of the two areas.
Origin of the Mid-Atlantic Ridge.—Of the many theories which have been proposed for the origin of the Mid-Atlantic Ridge almost all have been extremely speculative, and none has been based on any very detailed knowledge of the feature. We are still a long way from having a comprehensive knowledge of the Ridge. The various theories of origin and their factual basis have been briefly reviewed by Tolstoy and Ewing, who conclude that it is impossible to say if the feature is primarily of folded or faulted origin. In a paper in press Heezen and Ewing compare in detail the topography and seismicity of the African rift valleys and the Rift Valley of the Mid-Atlantic Ridge. Their conclusion is that the two areas are of basically the same structure, and in fact both form parts of the same continuous structural feature. Since the African rift valleys seem clearly to be the result of normal faulting resulting from extension of the crust, Heezen and Ewing conclude that the topography of the Mid-Atlantic Ridge is largely the result of normal faulting. Whether the forces are the result of horizontal extension or vertical uplift remains the most important unsolved problem in connection with the origin of the continental as well as the suboceanic rift-valley systems. Hess (1954) has proposed a mechanism relating suboceanic uplift to expansion due to serpentization of the upper mantle.