Timothy Reston

Detachment faulting during continental breakup and slow/ultraslow seafloor spreading

Detachment faults, apparently gently-dipping long offset normal faults, are observed in some continental rifts, at some rifted margins, and at both slow and ultraslow spreading ridges.  Key questions revolve around their angle and mechanics of slip: many papers have suggested that such detachments only ever slipped at high angles and have passively rotated to low angle.  In this talk,  I will discuss the mechanics of detachment faults at rifted margins, at slow- and ultra-slow spreading ridges.

At rifted margins, detachments include the S reflectors (Galicia and N Biscay margin), the P detachment (Porcupine Basin) and the H detachment (S Iberia Abyssal Plain).  The best constrained of these is the Galicia S reflector, imaged by a 3D seismic volume we collected in 2013.  The data have been processed through to prestack time migration, subsequently depth-converted, revealing the true 3D geometry of S and its relationship with the overlying block-bounding faults.  S is corrugated, confirming it is a slip surface: moving oceanwards the corrugations swing round from E-W  to ESE-WNW, suggesting that rifting may have migrated.  The faults above S die out and link laterally, requiring several faults to have ben active simultaneously: heave analysis allows us to group the faults into sets of contemporaneous structures.  Where the landward-most fault in each set detaches onto S, S is distorted as if the fault continues oceanwards as S, cutting across the more landward and older portions of S.  These observations suggest that S formed by some sort of rolling hinge mechanism, in which an apparently continuous detachment comprises the rotated root zones of successive normal faults, each cutting steeply through the hanging wall of the preceding, more landward fault.  However, the synrift geometries show that S must have slipped at ~20°, and as noted above, several faults (rather than a single fault) were active simultaneously.  We suggest that the latter is a requirement of the 3D nature of the world, and that the former indicates that S only developed once the entire crust had become brittle, allowing mantle hydration and the formation of weak (serpentine?) fault rocks.   IODP Proposal 943-Full sets out to test this interpretation, among other aspects of the rifting to breakup, by dating the synrift sequences across the margin.

To the south, a similar detachment (H) in the South Iberia Abyssal Plain is truncated oceanwards by another detachment that exhumes mantle.  The wide expanse of mantle that lies oceanward is itself characterized by landward-dipping reflections, which we interpret as the root zone of successive detachments, each of which moves asymmetrically with its hanging wall to be cut by a new fault cutting through its footwall.  The same model explains the unroofing of wide expanses of mantle during amagmatic spreading at the Southwest Indian Ridge; kinematic analysis suggest however that these fault must have initiated as very steep structures in marked contrast to S.  Seismicity also suggests that oceanic detachments, controlling the formation of oceanic core complexes (e.g 13°20’N, which we investigated during a cruise in 2016), also dip steeply (75-80°) at depth, and thus might have developed from dikes.  Thus there appear to be detachment and detachments active during rifting and seafloor spreading.