EVIDENCE AND MECHANISMS FOR FORE-ARC EXTENSION AT THE ACCRETIONARY COSTA-RICA CONVERGENT MARGIN

Seismic reflection data across the upper trench slope off the Nicoya P eninsula, Costa Rica, reveal a wide zone of nearly trench-parallel nor mal faults. Although work in the last decade has shown that normal fau lts are present at many convergent margins, most examples (e.g., Japan , Peru-Chile, and Guatemala) have been associated with margins experie ncing subduction erosion or non-accretion. In contrast, extension in t he Costa Rica study area apparently is coeval with frontal accretion a nd underplating. The normal faults across the Costa Rica forearc are s triking in seismic section due to the well-layered, 2-km-thick upper s lope apron. Fault plane reflections and reflector terminations show th at the faults extend through the sedimentary apron and apparently into the underlying accretionary prism, indicating a deep-seated deformati on process. The zone of extension is from the midslope area to within 10 km of the shelf edge, a minimum width of about 20 km; the estimated extension across the zone is at least 1.5 to 3 km. Within the apron s ection, spacing between the faults is generally 200-500 m, and nominal fault dip is 20-degrees-40-degrees and predominantly landward. Activi ty on the normal faults appears to have occurred over a significant pe riod of time based on increased displacement with depth and on fault-c ontrolled sedimentary thickening. At least some of the faults may be p resently active; shallow reflectors and possibly the seafloor are disp laced by faulting. Contemporary sediment accretion is documented by th e same seismic reflection profiles showing offscraping and underplatin g near the toe of the wedge and out-of-sequence thrusting primarily be low the midslope area. The consistent landward normal fault dip may be influenced by structural anisotropy in the prism and possible extensi onal reactivation of earlier thrust faults associated with accretion p rocesses. With the available data it is not possible to conclusively d etermine the cause of the stress field leading to the upper prism and apron extension. However, the three most likely causes are underplatin g, changes in basal shear stress, or a brief episode of subduction ero sion.