A new swath bathymetry compilation of the Gulf of Cadiz Area and SW Iberia is presented. The new map is the result of a collaborative research performed after year 2000 by teams from 7 European countries and 14 research institutions. This new dataset allow for the first time to present and to discuss the missing link in the plate boundary between Eurasia and Africa in the Central Atlantic. A set of almost linear and sub parallel dextral strike-slip faults, the SWIM Faults (SWIM is the acronym of the ESF EuroMargins project "Earthquake and Tsunami hazards of active faults at the South West Iberian Margin: deep structure, high-resolution imaging and paleoseismic signature") was mapped using a the new swath bathymetry compilation available in the area. The SWIM Faults form a narrow band of deformation over a length of 600 km coincident with a small circle centred on the pole of rotation of Africa with respect to Eurasia, This narrow band of deformation connects the Gloria Fault to the Rif-Tell Fault Zone, two segments of the plate boundary between Africa and Eurasia. In addition, the SWIM faults cuts across the Gulf of Cadiz, in the Atlantic Ocean, where the 1755 Great Lisbon earthquake, M∼8.5-8.7, and tsunami were generated, providing a new insights on its source location.

a b s t r a c tThe onshore sector of the West Iberian Margin (WIM) was the locus of several cycles of magmatic activity during the Mesozoic, the most voluminous of which was of alkaline nature and occurred between 70 and 100 Ma. This cycle took place in a post-rift environment, during the 35 counter-clockwise rotation of Iberia and initiation of the alpine compression. It includes the subvolcanic complexes of Sintra, Sines, and Monchique, the volcanic complex of Lisbon and several other minor intrusions, covering an area of approximately 325 km 2 . Previous cycles were tholeiitic and transitional in nature, occuring around 200 Ma and 130-135 Ma, respectively. New LA-ICP-MS U-Pb, 40 Ar/ 39 Ar, K-Ar and Rb-Sr ages on several intrusions distributed along the onshore WIM are presented, which combined with previously published data allows us to constrain the duration of the Late Cretaceous alkaline cycle to circa 22 Ma (94-72 Ma) and define two pulses of magmatic activity. The first one (94-88 Ma) occurred during the opening of the Bay of Biscay and consequent rotation of Iberia and clusters above N38 20 0 . The second pulse (75-72 Ma) has a wider geographical distribution, from N37 to N39 . This final pulse occurred during the initial stages of the Alpine orogeny in Iberia that led to the formation of the Pyrenees and Betics and to tectonic inversion of the Mesozoic basins. Isotope and trace element geochemistry point to a sublithospheric source for the alkaline magmatism that clearly distinguishes it from the previous cycles which had an important lithospheric mantle component. Also, it allows the discrimination between the two different alkaline pulses in terms of trace element abundance and residual mantle minerology. It is speculated that these differences might be the result of distinct magma ascent rates due to either more or less favourable tectonic settings that avoided or allowed the interaction with metasomatized lithosphere and equilibration with K rich minerals like amphibole and/or phlogopite.

Subduction initiation at passive margins plays a central role in the plate tectonics theory. However, the process by which a passive margin becomes active is not well understood. In this paper we use the southwest Iberia margin (SIM) in the Atlantic Ocean to study the process of passive margin reactivation. Currently there are two tectonic mechanisms operating in the SIM: migration of the Gibraltar Arc and Africa-Eurasia convergence. Based on a new tectonic map, we propose that a new subduction zone is forming at the SIM as a result of both propagation of compressive stresses from the Gibraltar Arc and stresses related to the large-scale Africa-Eurasia convergence. The Gibraltar Arc and the SIM appear to be connected and have the potential to develop into a new eastern Atlantic subduction system. Our work suggests that the formation of new subduction zones in Atlantic-type oceans may not require the spontaneous foundering of its passive margins. Instead, subduction can be seen as an invasive process that propagates from ocean to ocean.