Episodes of sea-floor spreading recorded by the North Atlantic basement

Abstract

A fresh compilation of available bathymetric, magnetic and seismic data in the central North Atlantic (west of the Mid-Atlantic Ridge) and north of the Azores (east of the Mid-Atlantic Ridge) reveals a complex episodic history of sea-floor spreading. JOIDES deep drilling results allow approximate dating of these episodes. By the end of the Jurassic, which was characterized by fast but decelerating spreading and few fracture zones, the central North Atlantic had reached about one third of its present width. Even before the end of this episode, close-spaced fracture zones and rough basement began to be formed, and this basement change appeared systematically later from north to south, consistent with the hypothesis that a threshold spreading rate was involved. As the spreading pole lay to the north, a decelerating spreading rate would necessarily bring a northern location through a given threshold rate before a southern one. Near the Jurassic-Cretaceous boundary, the Mid-Atlantic Ridge rapidly grew up to 0.5 km higher in some areas; this is known from basement topography and supported by a band of high-amplitude magnetic anomalies that suggest a thickened pillow lava layer about 50 km wide. This unusual band called the Bermuda Discontinuity, generally separates the rather densely fractured Cretaceous sea floor (about one fracture zone per 50–100 km) on the east flank of the Bermuda Rise from the Jurassic province to the west. Comparison of basement character with crustal structure suggests that, at least at the latitude of Bermuda, the “fast” Jurassic crust is “Pacific” (oceanic layer, normal mantle) whereas the Cretaceous sea floor is “Atlantic” (no oceanic layer, low velocity mantle). Sometime after the start of the Cretaceous, basement south of about 26° N again became “Pacific” (smooth, normal crust) whereas to the north it remained “Atlantic”. Cretaceous fractures yield a good azimuth to the virtual spreading pole, which is considerably southeast of the late Tertiary pole. The Cretaceous ended with the formation of basement elevations such as the Corner Rise. Major changes in spreading pattern probably occurred both at 76 and 60 m.y.B.P., the later date marking the separation of Greenland from Europe. Between about 60 and 40 m.y.B.P. fracture zones trend east-west or even WSW, suggesting the spreading pole moved west at the Cretaceous-Cenozoic boundary. After 40 m.y.B.P. large pole changes are not evident; a slight eastward movement may have occurred about 9 m.y.B.P. concomitant with the disappearance of numerous minor fracture zones. A preliminary comparison between spreading history on Reykjanes Ridge and deep drilling in Rockall-Hatton Basin suggests rapid subsidence following, by a few million years, each of three changes of spreading pattern. It is noteworthy that the three major events in the history of the North Atlantic all fall close to period boundaries, which therefore are not merely stratigraphic boundaries whose significance is restricted to Europe. Noting also the coincidence of the three major events with flood basalt episodes, we propose that discharge from mantle plumes, varying with crude 60 m.y. periodicity, has been the prime cause for the observed North Atlantic evolution.