Extension of continental crust at the margin of the eastern Grand Banks, Newfoundland

Abstract

Seismic and gravity observations from the rifted margin of the eastern Grand Banks, Newfoundland, support a new model for extension of the continental crust from the shelf edge to ODP Site 1277, where mantle rocks are exhumed. We find that the largest decrease in crustal thickness, from about 28 km to 6 km, occurs beneath the continental slope of the Grand Banks over a distance of just 20 km. This rapid decrease in crustal thickness coincides with anomalously high seismic velocities (7.0–7.2 km·s − 1 ) in the lower crust of the shelf edge. The thin crust of the continent–ocean transition (COT) in this area has a smooth basement surface, void of upper crustal blocks and prerift sediments. We compare our geophysical results with a geodynamical model that represents rifting of a relatively hot continental lithosphere and with another numerical model that represents rifting of a cold lithosphere. Both geodynamic models suggest that crustal thinning beneath the continental slope was achieved by extensional faulting in the upper crust and ductile shear zones in the middle crust. The geodynamic models provide an explanation for the formation of distinct continental slopes at rifted margins: Beneath the continental shelf of the Grand Banks, the Moho and the strong lower crust rotated upwards toward to a 50° dip without visible internal deformation. The presence of these strong lower crustal rocks at shallow depth in the rift flank subsequently helped to localize the extension farther seaward. With ongoing extension, some high-angle normal faults may have rotated to a sub-horizontal orientation, which would explain the lack of brittle deformation visible in the seismic reflection data. The two geodynamic models produce different amounts of extension of continental crust in the distal margins. The hot rifting model localizes strain much more rapidly, leaving narrow zones of extended continental crust, and it produces a relatively large amount of melt (> 30%) in the final stages of rifting. Continental breakup may occur rapidly in hot lithosphere (< 5 Myr). On the other hand, a cold extension model extends the continental crust to a thickness smaller than 10 km over a width of 50 km in the distal margin, similar to what we inferred at the eastern Grand Banks. The cold lithospheric model requires about 23 Myr of extension before continental breakup, and it predicts much less melting in the mantle (13%). The long rift duration, wide zones of thinned continental crust, and small amount of magmatism make the cold rifting model the most applicable to Newfoundland–Iberia rift.