A tectonic model for hyperextension at magma-poor rifted margins: an example from the West Iberia–Newfoundland conjugate margins

Abstract

AbstractHyperextended, magma-poor margins are characterized by a wide continent–ocean transition and anomalously small fractions of magmatism during mantle exhumation prior to oceanic spreading. Here, I bring together several aspects of their rift to drift transition and give a coherent picture of their evolution from platform to deep-sea environments. I focus mainly on the West Iberia Margin (WIM)–Newfoundland (NF) conjugates in the North Atlantic Ocean. The architectural evolution of these margins is characterized by upper-crustal faulting and lower-crustal deformation that are tightly coupled, resulting in fault displacement that is accompanied by underlying, equal and coeval crustal thinning. Lower crust deforms first ductilely, but then progressively switches to brittle due to enhanced conductive cooling at very slow extension velocities (<c. 6 mm a−1 half-rate). The switch from ductile to progressively brittle lower crust is accompanied by the emergence of a dominant basinwards fault dip and oceanwards younging of fault activity. It is shown that these processes, acting in concert: (1) reconcile the horizontal extension on faults with crustal thinning without the need of lower-crustal flow; (2) explain, within one common Andersonian framework (faults active at 65°–30°), the change in fault geometry from planar to listric to detachment-like with increasing extension; and (3) generate the tectonic asymmetry observed between conjugate pairs. This work also discusses a high-resolution seismic section of the WIM showing that the ‘detachment-like’ fault S is truncated prior to the peridotite ridge where mantle exhumation first takes place. This suggests that serpentinized mantle rises due to its own buoyancy, separating and pulling the thinned crustal blocks apart. Once the crust has been separated, further mantle exhumation takes place by magma-poor extension of the underlying mantle. I show with numerical models that either a small reduction of mantle potential temperature (c. 25–50 °C), a mantle depletion of more than 10% or very slow half-extension velocities (c. 6 mm a−1) are required to reproduce the small amount of magmatism inferred. Available observations support either a very slow extension velocity or a smaller than normal mantle temperature; however, estimation errors may be large. Ultimately, unravelling which of these factors most contribute to magma-poor mantle exhumation will provide an improved understanding of the mantle lateral homogeneity and the three-dimensional nature of the rifting to drifting process.