Mixed Brittle and Viscous Strain Localization in Pelagic Sediments Seaward of the Hikurangi Margin, New Zealand

Abstract

Calcareous‐pelagic input sediments are present at several subduction zones and deform differently to their siliciclastic counterparts. We investigate deformation in calcareous‐pelagic sediments drilled ∼20 km seaward of the Hikurangi megathrust toe at Site U1520 during International Ocean Discovery Program (IODP) Expeditions 372 and 375. Clusters of normal faults and subhorizontal stylolites in the sediments indicate both brittle faulting and viscous pressure solution operated at <850 m below sea floor. Stylolite frequency and vertical shortening estimated using stylolite mass loss, porosity change, and distribution increase with carbonate content. We then use U1520 borehole data to constrain a P‐T‐t history for the sediments and apply an experimentally derived pressure solution model to compare with strains calculated from stylolites. Modeled strains fail to replicate stylolite‐hosted strain distribution or magnitude, but comparison shows porosity, composition, and grain‐scale effects in diffusivity and mass transfer pathway width likely exert a strong influence on pressure solution localization and strain rate. Stylolite and fault clusters concentrate clay in these sediments, creating weak volumes of clay within carbonates, that may localize slip where the plate interface intersects the carbonates at <5‐km depth. Plate interface slip character and rheology will be influenced by the deformation of intermixed phyllosilicates and calcite, occurring by variably stable frictional slip and pressure solution of calcite. Pressure solution of calcite is therefore important at the shallow plate interface, waning at the base of the slow‐slipping zone because calcite solubility is low at temperatures >150°C where frictional (possibly seismic) slip likely predominates.

Frictional and Lithological Controls on Shallow Slow Slip at the Northern Hikurangi Margin

“…Broadly, our results indicate that mixing between the strong, brittle carbonates and the weak, viscous clay minerals could be an important two‐phase mineralogical control on healing and shallow SSE nucleation at the northern Hikurangi margin, consistent with structural observations by Leah et al. (2020). Because the shallow portion of the northern Hikurangi margin is relatively cold (Antriasian et al., 2019; McCaffrey et al., 2008), we do not consider the role of temperature here, but it could be an additional control at depth, particularly downdip of the seismogenic zone.…”

Plate‐Rate Frictional Behavior of Sediment Inputs to the Hikurangi Subduction Margin: How Does Lithology Control Slow Slip Events?

“…This cannot be explained by lithological differences or mineralogical changes due to diagenesis, because they are the same for both sample types, therefore these frictional differences must be the result of structural changes. Powdering the samples destroys the intact structure developed in‐situ, including porosity loss, grain orientations that have formed during compaction and also solution‐precipitation reactions (Leah et al., 2020). Powdering may also decrease the grainsize and destroy the nannofossils that are abundant in both tested samples.…”

“…Powdering the samples destroys the intact structure developed in-situ, including porosity loss, grain orientations that have formed during compaction and also solution-precipitation reactions (Leah et al, 2020). Powdering may also decrease the grainsize and destroy the nannofossils that are abundant in both tested samples.…”

Is complex fault zone behaviour a reflection of rheological heterogeneity?

“…If aseismic slip at the updip end of the seismogenic zone also involves some component of viscous deformation, then our assumption of a sharp, T -dependent onset of pressure solution at 100 ° C hides any shallower, near-trench effects of mixed frictional–viscous or seismic–aseismic deformation. Such shallow, visco-frictional deformation may occur in some subduction zones, particularly where carbonates make up a considerable proportion of the incoming sediment sequence [ 137 , 138 ]. We also recall that within the shallow, dominantly frictional regime, variations in consolidation, porosity, fluid pressure or material properties in time and space may add further heterogeneity than what we model here, even in the absence of a viscous component [ 47 ].…”