Abstract.We present a crust and mantle velocity structure for the West Iberia passive continental margin derived from a 320-kin-long wide-angle seismic profile acquired in the southern Iberia Abyssal Plain. We observe a 170-kin-wide oceancontinent transition zone which includes a pair of overlapping peridotire ridges and is bounded by

The rifting of continents involves faulting (tectonism) and magmatism, which reflect the strain-rate and temperature dependent processes of solid-state deformation and decompression melting within the Earth. Most models of this rifting have treated tectonism and magmatism separately, and few numerical simulations have attempted to include continental break-up and melting, let alone describe how continental rifting evolves into seafloor spreading. Models of this evolution conventionally juxtapose continental and oceanic crust. Here we present observations that support the existence of a zone of exhumed continental mantle, several tens of kilometres wide, between oceanic and continental crust on continental margins where magma-poor rifting has taken place. We present geophysical and geological observations from the west Iberia margin, and geological mapping of margins of the former Tethys ocean now exposed in the Alps. We use these complementary findings to propose a conceptual model that focuses on the final stage of continental extension and break-up, and the creation of a zone of exhumed continental mantle that evolves oceanward into seafloor spreading. We conclude that the evolving stress and thermal fields are constrained by a rising and narrowing ridge of asthenospheric mantle, and that magmatism and rates of extension systematically increase oceanward.

No abstract

Abstract. We present a wide-angle seismic refraction study of an 80x40 km region of the southern Iberia Abyssal Plain, south of Galicia Bank. An intersecting grid of two E-W and four N-S wide-angle reflection/refraction profiles is used to define variations of the basement velocity blocks immediately south of Galicia Bank. This crust is underlain by a high-velocity layer (7.3-7.9 km/s) of weakly serpentinized (i.e., 0-25%) peridotite, which exists throughout the eastern part of the survey area. Basement within the OCT appears to consist dominantly of a broad region of exposed upper mantle that has been serpentinized heterogeneously both vertically and horizontally. In the southeast sector of our survey where basement topography deepens and becomes subdued, continental fault blocks are absent; instead, basement contains an upper layer of more pervasively serpentinized (i.e., 25-45%) peridotite that is ~2 km thick. This layer is characterized by low velocity at the top of basement (4.2 km/s) that increases rapidly with depth, and it probably corresponds to a seismically unreflective layer, previously identified in reflection profiles to the south of our survey. In the western section of our survey, beneath a series of elevated basement ridges, velocities are reduced within both the upper •asement layer (3.5-6.0 km/s) and lower layer (6.4-7.5 km/s). These changes suggest that both upper and lower layers have become more highly serpentinized (with values of 60-100% in the upper layer and 25-45% in the lower layer) probably during the last stages of rifting and immediately before formation of oceanic crust. A normal or slow spreading oceanic crustal structure is not found within the survey region. Thus it appears that the onset of seafloor spreading occurs in the region west of the peridotite ridge sampled at ODP Site 897 and east of the J magnetic anomaly.

A seismic refraction transect across the Galicia Bank continental margin shows that the original continental crust thins westward from 17 to 2 km immediately east of a margin‐parallel peridotite ridge (PR). Immediately west of the PR, oceanic crust is only 2.5–3.5 km thick, but farther west (oceanward) it thickens to 7 km. The PR caps a ∼60‐km‐wide lens‐shaped serpentinized peridotite body underlying both thinned continental and thin oceanic crust. When superimposed on a reflection time version of the velocity model, the S reflector is clearly intracrustal at its east end. Westward, S cuts down to lower crustal levels, eventually coinciding with the top of the serpentinized peridotite lens (original crust‐mantle boundary). These observations render almost impossible the seafloor exposure of the PR by S acting as a top‐to‐the‐west detachment fault. Numerical models of melting and borehole subsidence information constrain our rifting model. The easternmost continental crust experienced a total stretching factor of 4.3 (most likely in two stages); it probably occurred over ∼25 m.y., with the highest rate of stretching at the beginning of the main earlier rift phase (Valanginian; 141–135 Ma). The 3 (4.7) km thick continental crust (depending on whether serpentinized peridotite is assigned to crust or mantle), which may include melt products, requires stretching factors of more than 11 (7) and a rift duration of more than 25 (13) m.y. The thin oceanic crust immediately west of the PR is explained by conductive cooling of the mantle during the long prebreakup stretching phase, which temporarily caused reduced melting immediately after breakup.