We thank Marques for his Comment (Marques, 2014) on our paper (Duarte et al., 2013) about subduction initiation at the southwest Iberia margin (SIM). Here we demonstrate that his comments are based on wrong premises:

(1) Marques ignores a fundamental premise of the setting of the studied area: the evidence for two tectonic driving mechanisms: the Gibraltar subduction and the oblique Nubia-Iberia convergence. Compressive stresses may arise from (a) an imbalance in horizontal forces due to partial slab breakoff and/or a decrease in the subduction rate; (b) expulsion of the Alboran block; (c) torques at the SIM; and (d) pull of the oceanic corridor toward Gibraltar, resisted by the adjacent northern oceanic segment (Gorringe area) being dragged against a relatively immobile Iberia.

(2) It is widely accepted that the Gibraltar arc formed due to subduction rollback. Evidence includes: 1) tomography showing a high-velocity east-dipping body up to the 660 discontinuity; 2) deep seismic events (>600 km); 3) the presence of a 15-km-thick accretionary wedge rooting in an east-dipping décollement); 4) mud volcanism in the foreland; 5) active thrusting in the wedge; 6) GPS data showing a westward-moving Alboran plate; 7) the existence of an oceanic corridor in the foreland; 8) existence of an extensional back-arc region; 9) calc-alkaline arc volcanism; and 10) pressure-temperature-time paths of high-grade metamorphic rocks in the Betics. The existence of subduction beneath Gibraltar was highly disputed in the past, but in the past decade the debate has shifted from “is there subduction in Gibraltar?” to whether the subduction is still active or not. In Duarte et al. (2013) we discuss both possibilities.

(3) Lithospheric-scale structures cannot be directly deduced from crustal seismics. This is a limitation of the method, and not evidence that the structures are too small to be lithospheric. Complementary methods need to be used, such as earthquake hypocenters. In the studied area, the seismicity is concentrated at ∼40–60 km depth, with focal mechanisms showing reverse faulting and NW-SE shortening (Geissler et al., 2010). Hence, it is reasonable to infer that the compressive structures root into the mantle.

(4) We referred the 1969 and 1755 earthquakes because their high magnitudes (∼8–9 Mw) are characteristic of convergent zones. And even though the location of the 1755 quake is uncertain, the source of the 1969 quake lies 50 km southeast of the Gorringe, in the flat abyssal plain. The fault plane solution indicates conjugate ∼N60°-striking 45°-dipping planes with a reverse motion. Fukao (1973) favored the southern conjugate based on the distribution of the aftershocks, but at the time there was a large uncertainty in their location. No major northwest-dipping structures exist in this area. In our map, we show a set of southeast-dipping thrusts that seem to be connected and could have generated the 1969 quake. When Fukao published his paper, the Gorringe root was interpreted as a northwest-dipping megathrusts, with “consumption of oceanic lithosphere” (Purdy, 1975, p. 973). In the models of Purdy (1975) and McKenzie (1977), the African plate was subducting beneath Eurasia (the subduction zone was part of the east-west Azores-Gibraltar plate boundary) and what we propose is the nucleation and northward propagation of a new (convergent) plate boundary along the SIM.

(5) The main idea of our paper is that the interference between the Gibraltar subduction zone and the Azores-Gibraltar Fault Zone can seed the reactivation of the SIM. Calculation of the forces involved in this process is complex and beyond the classical 2-D evaluation of buoyancy versus ridge push forces used for spontaneous subduction initiation. Three-dimensional modeling (analog/numerical) would be necessary to understand the forces involved in such case of induced initiation. Two-dimensional models of spontaneous initiation are inadequate to quantitatively gauge subduction initiation along the SIM. Nevertheless, estimation of the Gibraltar slab pull force and the force required to break a passive margin are provided.

(6) Marques states “subduction in northern Iberia is overlooked … . By doing so, the evaluation of the force balance along the SIM is flawed.” The northern Iberia incipient accretionary wedge formed due to the closing of the Tethys, and has been inactive since the Burdigalian (ca. 20 Ma). Thus it is irrelevant to the present-day force balance. Marques further argues “overthrusting in northern Iberia could take up all the Nubia-Iberia convergence … , which makes it critical in the evaluation of the force balance in Iberia.” At present, this is not the case. The observations show that shortening is being accommodated at the SIM.

Marques’ main arguments are: (1) lack of “relevant data”; (2) absence of “use [of] critical model results that would constrain their [our] hypothesis of subduction initiation”; and (3) overlooking “relevant literature.” But Marques fails to explain exactly what relevant data are missing, especially when a new tectonic map was presented to sustain our hypothesis. The second point is based on the wrong premise that 2-D models of spontaneous subduction initiation can be used to constrain this case of induced initiation, whereas a 3-D approach would be required. Consequently the claim that we “overlooked relevant literature” is mostly unfounded.