Oliver [1992] has argued that during continental collision, overthrusting and tectonic shortening can cause compaction/tectonically induced groundwater flow. The existence of petroleum and metal ore deposits at the margins of continental basins far from the tectonic belt is frequently cited as supporting evidence for the ``squeegee'' model. However, the estimated magnitude of groundwater flow rates due to sediment compaction associated with overthrusting [ Deming et al., 1990] and tectonic shortening [ Ge and Garven, 1992] would only produce modest temperature anomalies. Alternatively, during orogenic events, uplift of the land surface and water table can induce regional topography-driven subsurface flow systems [ Sverjensky and Garven, 1992]. Numerical models of topography-driven groundwater flow predict significant thermal and salinity anomalies at the margins of basins consistent with observational data. However, this has been disputed by Deming and Nunn [1991].
Within subduction zones, recent efforts to document fluid migration have focused on the hydrodynamics of accretionary prisms [ Screaton et al., 1990] and the dehydration of minerals in the descending slab during prograde metamorphism [ Peacock, 1990; Elderfield et al., 1990]. Fluid flow in the accretionary wedge environment is induced by pore space collapse due to mechanical shearing and loading as sediments are scraped off the subducting oceanic plate (Fig. 1). Sediment compaction can also be induced by tectonic compression and thrust sheet loading. Fluids near the base of the subducting oceanic crust are driven upwards towards the ocean-sea floor interface. Fluids can also be driven laterally across the length of the accretionary prism if permeable layers are present. Compaction-driven fluid flow has been proposed as the most viable mechanism to explain the occurrence of fluid vents and anomalous methane fluxes at the base of the prisms [ Screaton et al., 1990; Oliver, 1992]. Subducting oceanic crust also produces large quantities of fluids [ Peacock, 1990]. Hydrated oceanic crust and the remaining sediments devolatize on subduction. Water rich fluids produced in the slab [ Peacock, 1990] infiltrate and metasomatize the overlying mantle wedge [ Vidal et al., 1989]. Due to the resulting high water activity, partially melting of the overlying mantle and crust occurs [ Hawkesworth et al., 1991]. The ascending magmas carry significant amounts of fluid [ Bergantz, 1991], which are expelled upon cooling. Numerical modeling studies indicate that local heat disturbances caused by magma bodies intruding the crust can generate vigorous density-driven flow systems [ Furlong et al., 1991, Hanson, 1992]. Locally, boiling can occur [ Ingebritson and Scholl, 1993] resulting in hydrologic brecciation of the host rock.
Within extensional tectonic settings, groundwater flow within
fault-bounded, continental rift basins can be induced by a number
of mechanisms including seismogenic pumping [ Sibson,
1994], compaction-driven [ Harrison and Summa, 1991],
density-driven [Raffensperger and Garven, in press], and
topography-driven [ Person and Garven, 1992; 1994] flow.
Under marine incursion or low topographic relief, free convection
can occur within thick (> 1km), permeable (> 10
m
) basal aquifers [Raffensperger and Garven, in press]. In
rapidly subsiding continental rifts or passive margins, such as the
Gulf Coast, USA [ Harrison and Summa, 1991], excess
pressures can approach lithostatic levels. Salinity effects can be
especially important within continental rifts which are
hydrologically closed [ Duffy and Al-Hassan, 1988] or contain
salt domes [ Evans et al., 1991]. However, it is unclear from
recent field observations whether salinity plumes above salt domes
are due to convection or upward flow of compaction-driven fluids
episodically released along fault zones. Along basement detachment
faults, which are thought to extend to the base of the continental
crust (Fig. 1), episodic dilation of pore spaces along extensional
faults and shear zones can focus crustal fluids at great depth [
Sibson, 1994]. Some studies have proposed seismogenic pumping
as a possible mechanism for the formation of low-angle normal
faults, the chemical alteration of sediments [ Bartley and
Glazner, 1985] and rift related ore genesis [ Sibson et al.,
1988]. While conceptual models of seismogenic pumping indicate
that pore fluids are focused upwards, its role in basin hydrology is
poorly understood.