Segment-Scale Seismic Structure of Slow-, Intermediate-, and Fast-Spreading Mid-Ocean Ridges: Constraints on the Origin of Ridge Segmentation and the Geometry of Shallow Mantle Flow

Abstract:

We present a synthesis of seismic observations that allow us to test models of ridge segmentation and subridge mantle flow beneath active oceanic spreading centers. Active-source experiments conducted along segments of the slow-spreading Mid-Atlantic Ridge (MAR), the intermediate-spreading Juan de Fuca Ridge (JdFR), and the fast-spreading East Pacific Rise (EPR) resolve the seismic velocity structure of the crust and uppermost mantle on the scale of 5 to 10 kilometers. Second-order discontinuities in the mid-ocean ridge (MOR) system have commonly been attributed to a decrease in magma supply from the upwelling mantle. However, the observed increase in the magnitude of mantle low-velocity zones, interpreted as zones of subcrustal melt accumulation, near these discontinuities is in direct contradiction to this view. Segment-scale measurements of seismic anisotropy at these three locations also reveal a new characteristic of MORs that may be commonplace: A systematic misalignment between the orientation of plate divergence and the direction of mantle flow. Crustal anisotropy, assumed to result from preferentially aligned cracks, is used to infer the current orientation of tensile tectonic stress driving plate separation. At the MAR, JdFR, and EPR the orientation of crustal anisotropy is consistent with the orientation of plate divergence predicted from plate velocity models. Passive mantle flow models predict that mantle anisotropy, resulting from strain-induced olivine LPO, should trend parallel to the direction of plate divergence. However, beneath each of these spreading segments, mantle anisotropy is rotated away from the plate divergence direction. The geometry of these misalignments correlates with recent rotations of the Euler poles in each plate system and with the sense of offset at 2^nd-order discontinuities. We discuss alternative models of ridge segmentation and mantle flow geometries and their implications for lithosphere-asthenosphere coupling.