Subduction Geometries in Northwestern South America
By Carlos Alberto VARGAS
Pages 225–245
Manuscript accepted October 29, 2018
Abstract
Using hypocentral solutions and arrival times of first P and S waves recorded by stations of the National Seismological Network of Colombia (NSNC), as well as GPS, gravity and magnetic public datasets, I have estimated velocity tomograms, Curie depth points and the strain field along NW South America to evaluate the subduction process and interactions of the first–order tectonic blocks. A wire model has been estimated supported by three profiles based on gravity forward modeling, earthquake distribution and b–values to elucidate the subduction behavior of the Caribbean and Nazca Plates under the South America Plate, highlighting at least three subduction scenarios, where in addition to the Caldas lithospheric tear, other minor tears are found in the lithospheric system of this region. Although it is possible a flat subduction along NW Venezuela, it is presented as an alternative hypothesis a steeper subduction, which mechanically is coherent with the structural features observed in this region. The wire model shows how the Caribbean Plate accommodates mechanically to change from flat subduction in the south to steeper subduction in the north, differentially uplifting the Santa Marta and Santander Massifs along a weakness zone that corresponds to the Santa Marta–Bucaramanga Fault System. The absence of a modern volcanic arc in the Eastern Cordillera and/or the Perijá range is a consequence of slow low–angle subduction, which is associated with the compressional regime induced by the Panamá tectonic indenter. In this scenario, I hypothesize the presence of a zone of fluid accumulation (> 130 km depth) derived from the dehydration process; these fluids cannot ascend to the surface, which impedes the formation of current active magmatism. However, during the last 9–12 Ma of relevant influence of the Panamá Arc against NW South America, other emplacements of magmatic material might have occurred along this orogenic system. The wire model also shows that the low seismic activity within the Antioquia Batholith is a consequence of its rigidity, promoting the transfer of strain derived from the subduction process from west to east, generating high seismic activity along its borders and suggesting that compositional and elastic properties at depth maintain its coherence as a structural body beyond the upper crust. A similar interpretation is indicated for the southern Eastern Cordillera.
Keywords: local earthquake tomography, Curie point depth, strain field, subduction, Caribbean Plate, Nazca Plate.