The ocean/continent transition beneath the Iberia Abyssal Plain and continental-rifting to seafloor-spreading processes

Abstract

The west Iberia continental margin is a nonvolcanic rifted margin which, following rifting that began in the Late Triassic, broke away from Newfoundland in the Early Cretaceous as rifting propagated from south to north. The ocean/continent transition off Newfoundland seems to occur ~50 km seaward of the shelf edge near the base of the continental slope. Off west Iberia, the ocean/continent transition is defined by an 80- to 130-km-wide region beneath the Iberia Abyssal Plain between the most oceanward-tilted basement fault blocks (continental crust?) and a 300-km-long narrow peridotite ridge parallel to the margin. Leg 149 drilled a west-to-east transect of holes across the ocean/continent transition, and three of these reached acoustic basement, beginning at the peridotite ridge. Sites 897 (over the peridotite ridge) and 899 (19 km further east) sampled serpentinized peridotites with similar, but not identical, petrologies that experienced a similar history of exhumation from the deep mantle to the surface. They crystallized at 1170°-1230°C, and this was followed by ductile shear deformation at 880°-1000°C. After limited partial melting, secondary minerals crystallized at about 30 km depth. Mylonitization, and then low-temperature deformation and serpentinization, followed as the rocks were exhumed at the seabed. One possibly important, but unexplained, difference between the two sites is the fivefold greater strength of remanent magnetization of the Site 899 cores, which is reflected in the amplitudes of magnetic anomalies observed over the sites. The peridotites are neither clearly subcontinental nor suboceanic. Further landward, Site 900 sampled a flasered cumulate gabbro basement. The rare-earth element (REE) patterns of the gabbros are ambiguous and have been matched by different authors to both island arc and N-MORB basalts. The light REE patterns fit a transitional MORB parent magma whereas Nd and Pb isotope ratios strongly suggest a MORB parent. The gabbro therefore likely formed in a magma chamber from a melt that was little contaminated, if at all, by continental crust. Traces of the primary mineralogy indicate dynamic crystallization at depths of at least 13 km and temperatures typical of granulite facies conditions. During exhumation to the seabed the rocks underwent retrograde metamorphism at 280 ± 20°C followed by crystallization of some plagioclases at 136.4 ± 0.3 Ma. Late Barremian- to late Aptian-age debris-flow and mass-flow deposits were encountered above acoustic basement at Sites 897 and 899, both of which are now situated on substantial basement elevations hundreds of meters above basement of the adjacent basins. They are overlain by poorly fossiliferous uppermost Maastrichtian to Eocene thin claystones and conglomerates (lag deposits) of similar age to the sediments at the same level in the flanking basins. We explain these deposits by rapid Aptian uplift (or relative uplift), about 10 m.y. after the onset of 10 mm/yr seafloor spreading west of the peridotite ridge, followed in Eocene time by blanketing by the sediments of the Iberia Abyssal Plain. This also explains the long history of seawater alteration of serpentinized peridotite at both sites. The above drilling results are explained here by two possible hypotheses, that also take account of geophysical constraints on the development of the west Iberia ocean/continent transition, neither of which explains all the observations. The first hypothesis is that the ocean/continent transition is underlain by crust formed by ultraslow (~5 mm/yr half-rate) seafloor spreading which, by analogy with the slow-spreading Mid-Atlantic Ridge, leads to the seabed exposure of peridotite and gabbro by extensive faulting. This hypothesis explains the drilling results but has difficulty in explaining the magnetic anomalies and the minute volume of basalt encountered in the cores. The second hypothesis envisages an ocean/continent transition of tectonically and magmatically disrupted continental crust. It explains the Site 900 MORB gabbro as material underplated beneath thinned continental crust or as an aborted point-of-initiation of seafloor spreading, and the alkaline to transitional characters of igneous clasts in Site 897 and 899 deposits as evidence of continental rifting. Its main problem is the lack of unambiguous continental basement samples although there is evidence of re-worked continental sediments in the cores. Both hypotheses predict a highly heterogeneous crust, which will be hard to characterize without further basement sampling.