Crustal and upper mantle seismic structure beneath the rift mountains and across a nontransform offset at the Mid‐Atlantic Ridge (35°N)

Abstract

We present new results on the crustal and upper mantle structure beneath the rift mountains along two segments of the Mid‐Atlantic Ridge and across a nontransform offset (NTO). Our results were obtained from a combination of forward modeling and two‐dimensional tomographic inversion of wide‐angle seismic refraction data and gravity modeling. The study area includes two segments: OH‐1 between the Oceanographer fracture zone and the NTO‐1 at 34°35′N and OH‐2 between NTO‐1 and the NTO at 34°10′N. The center of OH‐1 is characterized by anomalously thick crust (∼8 km) with a thick Moho transition zone with Vp = 7.2–7.6 km/s. This transition zone, coincident with a gravity low, is probably composed of gabbro sills alternating with dunites, as observed in some ophiolites. OH‐1 has larger along‐axis crustal thickness variations than OH‐2, but average crustal thicknesses are similar (6.0±1.2 km at OH‐1, 6.1±0.7 at OH‐2). Thus we do not find significant differences in magma supply between these segments, in contrast to what has been inferred from morphological and gravity studies. At both segments the shoaling of the Moho is more rapid at the inside than at the outside corners, consistent with models in which the inside‐corner crust is technically modified. The structural differences between inside‐ and outside‐corner crust are more apparent at OH‐2, suggesting that the extrusive layer is thinner at the inside corner of OH‐2 than at the inside corner of OH‐1, probably due to differences in axial morphology and along‐axis magma transport. NTO‐1 is characterized by a nearly constant velocity gradient within the upper 5 km and low upper mantle velocities (7.4–7.8 km/s). The anomalous structure beneath NTO‐1 is interpreted as fractured mafic crust. The P wave velocities and densities required to match the gravity data suggest that serpentinites are common beneath the NTO‐1 and possibly beneath the inside corners. Serpentinization could be as much as 40% at ∼3.8 km below seafloor and probably does not occur at subseafloor depths greater than ∼6.2 km at the NTO‐1. Our results indicate that in a slow spreading environment where magmatism and tectonism are equally important, the seismic Moho cannot be correlated with an unique geological structure. At the center of a segment the seismic Moho may represent the lower boundary of an interlayered grabbro‐dunite transition zone, while beneath the inside corner and NTO where the crust is thinner, it may correspond to an alteration front.