SYNCONVERGENT EXTENSION IN THE PERUVIAN ANDES ABOVE A MODERN-DAY FLAT SLAB

Active extensional faults are common in the Peruvian Andes above the modern-day flat Nazca slab. A crustal transect across the central Peruvian Andes at 9°S latitude shows a series of large-scale, west-dipping normal faults, including the Cordillera Blanca detachment fault, the Quiches fault, and a series of normal faults to the east in the Maranon massif. At least two of these faults are deep-seated, with seismicity to ≥15 km. Together these faults mark a very pronounced west-dipping structural asymmetry in the upper crust. Extension along the most spectacular of these faults - the Cordillera Blanca detachment fault - has unroofed the Late Miocene (~8 to 5.3Ma) Cordillera Blanca batholith, which now forms the backbone of the highest mountain range in Peru. Significantly, many of the structural features observed in the Cordillera Blanca and Maranon regions resemble those seen in young metamorphic core complexes. The central Andean region of Peru is thus currently dominated by extensional deformation, with contraction isolated to the trench and sub-Andean regions.

Although extension was enhanced locally by topographic and thermal effects (e.g., emplacement of Cordillera Blanca batholith), these factors alone do not adequately explain the extent, structural asymmetry and deep-seated nature of extensional faults in Peru. We suggest that subduction of the aseismic (hotspot) Nazca Ridge - and subsequent flattening of the Nazca slab - played a key role in driving extension in Peru. Plate reconstruction shows that subduction of the Nazca Ridge under the Cordillera Blanca region preceded both the cessation of arc magmatism (ca. 5.3Ma) and the onset of detachment faulting (ca. 5.4Ma). Strong spatial and temporal ties thus exist between aseismic ridge subduction, arc extinction due to slab flattening, and the initiation of detachment faulting. We propose that lithospheric buoyancy from the subducted ridge triggered extensional collapse of pre-thickened continental crust, with large-scale detachments nucleating in upper-crustal regions thermally weakened by batholith emplacement. The model requires that mid-to-upper crustal strain be coupled to the plate interface via a strong lower crust, plausibly the case in Peru given refrigeration of overriding South American lithosphere by flat slab subduction.