Description: Water and carbon are transferred from the ocean to the mantle in a process that alters mantle peridotite to create serpentinite and supports diverse ecosystems1. Serpentinized mantle rocks are found beneath the sea floor at slow- to ultraslow-spreading mid-ocean ridges1 and are thought to be present at about half the world’s rifted margins2, 3. Serpentinite is also inferred to exist in the downgoing plate at subduction zones4, where it may trigger arc magmatism or hydrate the deep Earth. Water is thought to reach the mantle via active faults3, 4. Here we show that serpentinization at the rifted continental margin offshore from western Spain was probably initiated when the whole crust cooled to become brittle and deformation was focused along large normal faults. We use seismic tomography to image the three-dimensional distribution of serpentinization in the mantle and find that the local volume of serpentinite beneath thinned, brittle crust is related to the amount of displacement along each fault. This implies that sea water reaches the mantle only when the faults are active. We estimate the fluid flux along the faults and find it is comparable to that inferred for mid-ocean ridge hydrothermal systems. We conclude that brittle processes in the crust may ultimately control the global flux of sea water into the Earth.

Description: Slow slip events (SSEs) at the northern Hikurangi subduction margin, New Zealand, are among the best-documented shallow SSEs on Earth. International Ocean Discovery Program Expeditions 372 and 375 were undertaken to investigate the processes and in situ conditions that underlie subduction zone SSEs at the northern Hikurangi Trough. We accomplished this goal by (1) coring and geophysical logging at four sites, including penetration of an active thrust fault (the Pāpaku fault) near the deformation front, the upper plate above the SSE source region, and the incoming sedimentary succession in the Hikurangi Trough and atop the Tūranganui Knoll seamount; and (2) installing borehole observatories in the Pāpaku fault and in the upper plate overlying the slow slip source region. Logging-while-drilling (LWD) data for this project were acquired as part of Expedition 372, and coring, wireline logging, and observatory installations were conducted during Expedition 375. Northern Hikurangi subduction margin SSEs recur every 1–2 y and thus provide an ideal opportunity to monitor deformation and associated changes in chemical and physical properties throughout the slow slip cycle. In situ measurements and sampling of material from the sedimentary section and oceanic basement of the subducting plate reveal the rock properties, composition, lithology, and structural character of material that is transported downdip into the SSE source region. A recent seafloor geodetic experiment raises the possibility that SSEs at northern Hikurangi may propagate to the trench, indicating that the shallow thrust fault (the Pāpaku fault) targeted during Expeditions 372 and 375 may also lie in the SSE rupture area and host a portion of the slip in these events. Hence, sampling and logging at this location provides insights into the composition, physical properties, and architecture of a shallow fault that may host slow slip. Expeditions 372 and 375 were designed to address three fundamental scientific objectives: Characterize the state and composition of the incoming plate and shallow fault near the trench, which comprise the protolith and initial conditions for fault zone rock at greater depth and which may itself host shallow slow slip; Characterize material properties, thermal regime, and stress conditions in the upper plate directly above the SSE source region; and Install observatories in the Pāpaku fault near the deformation front and in the upper plate above the SSE source to measure temporal variations in deformation, temperature, and fluid flow. The observatories will monitor volumetric strain (via pore pressure as a proxy) and the evolution of physical, hydrological, and chemical properties throughout the SSE cycle. Together, the coring, logging, and observatory data will test a suite of hypotheses about the fundamental mechanics and behavior of SSEs and their relationship to great earthquakes along the subduction interface.

Description: We conducted discrete numerical simulations to examine the effects of seamount collisions with forearcs along actively accreting subduction margins. Modeled seamount interactions leave behind distinctive structures in overriding forearcs that differ from those found at non-accreting margins. Whereas accretion above a planar décollement produces evenly spaced thrust faults with uniform displacements, seamounts activate one or more large-offset splay faults that accommodate substantial offset. Locally oversteepened slopes develop above the seamounts, but in contrast to non-accreting margins, the steep slopes are transient. Renewed accretion following seamount passage allows the equilibrium surface slopes to recover. Seamounts also protect incoming strata in their wake, delaying formation of new thrust faults and increasing fault spacing. Weak horizons within accreting strata allow the décollement to step up above the seamount, further protecting deeper strata and vertically partitioning wedge deformation. Notably, all modeled faults form in sequence, in contrast to out-of-sequence faults found at non-accreting margins. Similar structures found at many accretionary margins, including Nankai (offshore Japan), suggest that we may underestimate the role of seamount interactions in many locations, with implications for our assessment of subduction hazards in these settings.