Could Fluid Seeps Originate from the Seismogenic Zone? Evidence from Southern Costa Rica

The prevailing conceptual model of convergent margin hydrogeology is one in which fluid sourced from porosity loss and dehydration reactions seaward of the updip limit of the seismogenic zone reach the seafloor via relatively low angle splay faults that act as high permeability conduits through an otherwise nearly impermeable upper plate [e.g., Lauer and Saffer, GRL, 39:L13604, 2012; Saffer and Tobin, Ann. Rev. Earth Planet. Sci., 39:157-186, 2011]. Interpretation of newly acquired 3D seismic reflection data and high resolvability multibeam and backscatter data, showing evidence for abundant potential fluid seeps sourced beneath the sediment cover and farther landward than previously thought possible, may require reevaluation of this concept. Kluesner et al. [2013, G3, doi:10.1002/ggge.20058], identified 160 potential fluid seeps in an 11 km wide swath off southern Costa Rica, based on pockmarks and high backscatter mounds, each showing subsurface indicators of fluid migration in the seismic data. Approximately half of these potential seeps are on the outer continental shelf; these are landward of the updip limit of the seismogenic zone, as estimated by both the transition from high to low reflectivity of the plate boundary and the intersection of the 150°C isotherm with the plate boundary [Ranero et al., 2008, G3, doi:10.1029/2007GC001679; Bangs et al., 2012, AGU Fall Meeting, T13A-2587; Bangs et al., this meeting]. We have mapped high probability fluid pathways beneath these potential seeps, based on seismic meta-attribute volumes calculated using user-trained neural network algorithms [Kluesner et al., this meeting]. The mapped fluid pathways are high-angle through the sedimentary section, and they root into basement highs and basement faults. Fluids could originate along the plate interface, where potential sources and pathways are known (Mid-slope sites: Hensen et al., 2004, Geology, 32:201-204), or above or below the interface, although sources from these regions have not been reported. They could travel near vertical paths through the crustal rocks, or along a landward-dipping path, because the seismic data show landward dips but not seaward dips. If the fluids do come from the plate interface, they originate in the seismogenic zone. This inference can be tested by geochemical study of the outer shelf fluid seeps, where such sampling has not yet occurred.