Fluid flow and coupled processes at active margins: Case studies for the Barbados and Costa Rica subduction zones

Circulation of fluids beneath the seafloor is of great interest, particularly in subduction zones, where fluids may facilitate slip along the plate interface by reducing shear strength and effective stress. In turn, the earthquake cycle may enhance fluid flow and heat transport through the crust. Sub-seafloor fluid circulation and fault-valve mechanisms are commonly invoked to explain thermal anomalies in tectonically active areas. Despite the significance of these processes, the temporal relationships among stress, strain, and crustal fluids throughout the earthquake cycle, as well as the driving mechanism for fluid flow following earthquakes, remain poorly understood.; This research investigates how faulting may influence fluid flow and heat transport at active convergent margins. Two and three-dimensional numerical models were applied to two subduction-zone settings: Barbados and Costa Rica. The modeling was constrained by seismic reflection, hydrogeologic and geothermal data. The analyses quantified the effects of permeability enhancement and coseismic strain on fluid flow and heat transport.; A coupled three-dimensional numerical fluid flow and heat transport model was developed to explore the effect of permeability enhancement on fluid flow and advective heat transport at the Barbados accretionary complex. The thermal response of the system to enhanced fault permeability indicated that the effects of episodic fluid flow and advective heat transport were far more significant than steady-state flow and transport produced by the along-strike change in thickness of the accretionary prism. A two-dimensional steady-state model was also created for the Costa Rica margin. Fluid flow and heat transport were coupled with an earthquake strain model to evaluate how seismically induced poroelastic deformation and resultant fluid flow affected fluid pressures and temperatures within the Costa Rica subduction zone. Patterns of pore-pressure recovery were more variable than those predicted by theoretical faulting models. Coseismic compression and extension of the crust produced high fluid pressures close to the fault, whereas the inflow of fluid from depth increased fluid pressures for several years following the simulated fault slip. Crustal deformation alone was not observed to perturb the temperature field. Localized or laterally extensive permeability changes of up to two orders of magnitude, were required to produce heat flow anomalies.