Fluid flow at the toe of the Barbados accretionary wedge constrained by thermal, chemical, and hydrogeologic observations and models

Abstract

Cores and geochemical observations from the plate boundary (decollement zone) at two sites on the Barbados wedge indicate the occurrence of episodic fluid flow along the decollement, but subsequent CORK (Circulation Obviation Retrofit Kit) experiments (pressure and temperature monitoring in cased and sealed holes) provided no evidence for temperature transients associated with fluid flow and measured pressures that indicate moderate fluid overpressuring. Comparison of observations and thermal models of steady state and transient fluid flow along the decollement suggests that the decollement has not been the main conduit for fluid flow for the last few thousand years. However, pore fluid chemistry indicates recent, and possibly still active, flow along the decollement. Comparison of results from advection‐diffusion models applied to the chloride anomaly observed at the top of the decollement at Site 948/671 indicates that fluid flow may have been continuous over the last 2000–20,000 years, but mineral vein structure and chemistry suggest that episodes of higher pore pressure and water flux also occurred in the past. In situ permeability measurements show a strong dependence of permeability on pore pressure. A decollement model taking into account this dependence suggests that the propagation of a pressure front associated with a flow pulse could occur with characteristic times compatible with those inferred from the chlorinity profile. In a wedge cross section the fluid flux associated with a pressure pulse is probably <100 m2 yr−1 and thus may not explain heat flow anomalies observed near the toe of Barbados wedge. Thermal anomalies are better explained by fluid flow within the underthrust sequence and, perhaps, in basement.