Upper-crustal Stress Field Variations During the Building of the Central Andes: Constrains on the Activation/deactivation of Megadetachments

Abstract:

Nowadays, it is broadly accepted that the Central Andes resulted largely from crustal shortening in the last ~45 Ma, driven by horizontal forces as a consequence of subduction of the Nazca plate beneath South America. However, the way this shortening is achieved is still a matter a debate. Structural, seismological, thermochronological, isotopical and sedimentological studies of the Central Andes, together with thermomechanical modeling, suggest that different megadetachments located shallow in the upper crust were active during the construction of the Andes. Constrains on changes in the state of stress in the crust gleaned from more than 1,500 fault-slip data in the arc region provide insights into how and when these megadetachments get activated or deactivated. We used a forward modeling procedure to examine five transects across the Central Andes, at 21.5°, 24°, 30°, 34° and 35°S, with particular emphasis on the relationship between deep and shallow structures. Our kinematic-thermomechanical models show that most of the upper-middle crust has a brittle-elastic behavior particularly for the cold and rigid forearc and foreland regions, and a ductile behavior below the thermally weakened arc region. Our models assume a shallow, sub-horizontal megadetachment located at the shallowest brittle-ductile transition, which concentrates the majority of the horizontal crustal shortening between the fore-arc and the South American craton. During this horizontal shortening, the crust gets thick and topography rises due to buoyancy of the crustal root. The threshold of this thickening is achieved when the bouyancy force equals the horizontal force. At this point, the megadetachment deactives and the crustal root widens eastwards in concert with ductile deformation in the lower crust and the generation of a new megadetachment. By studying changes in the paleostress fields along the arc region, from compression to strike-slip, and strike-slip to extension, associated with σ₃/σ₂ and σ₂/σ₁ permutations respectively, together with the timing of uplift and exhumation of the morphostructural units across the transects, we can constrain the timing of activation/deactivation of the detachments responsible for the Andean deformation.