Item

Abstract

Combined geochemical and structural analyses are used to estimate the relative proportions of both tectonic thickening and addition of underplated magmas in a transpressional orogen. Granitoid intrusions were emplaced prior to and contemporanous with crustal shortening in the Late Cretaceous to Late Eocene paleo-arc in northern Chile (~22°S). Unlike the presently active volcanic arc, this partly eroded paleo-arc (up to 3 km) allows to quantify crustal shortening and estimate the proportions of volcanic and intrusive rocks (7.7:1), at least at the crustal level exposed. Temporal changes in the major and trace element patterns of these rocks show a gradual transition from lower pressure (gabbroic) to higher pressure (amphibolitic) residual mineralogies prior to, and the onset of garnet immediately after crustal shortening. This is generally attributed to deepening of the melt source region at or near the crust-mantle boundary. Gabbroic residual mineralogies are consistent with crustal thicknesses of ~37 km (thickness of the Brazialian shield), whereas the onset of residual garnet may reflect crustal thicknesses of ~45 km. Crustal thicknesses exceeding 50 km are restricted to residual eclogite; this is not seen in the Eocene melts. Temporal and spatial constraints on late Eocene crustal shortening come from detailed field studies and transpressional kinematic modeling. Balanced structural cross-sections indicate 9 km of horizontal shortening. Two-step transpressional deformation model involving pure shear (vertical extrusion and horizontal shortening) and subsequent simple shear in a vertical plane suggests crustal thickening by 14.5% to ~42 km crustal thickness. Consistent with experimental petrological work and results from geochemical modeling, we argue that the remaining 3 km (or 6.2%) of crustal thickening (to 45 km) were achieved by underplating mantle-derived basalts at or near the base of the arc crustal base prior to and during transpression. The resulting ratio of tectonic to magmatic crustal thickening lies therefore in the order of ~2:1. The proposed thickness of accreted mafic crust (~3 km) lies well within the potential range predicted for magmatic underplating (up to 20 km), and the resulting crustal addition-rates during the ~12 m.y. of arc transpression (35.5 km3/km) are well within the range of that proposed for other Phanerozoic arcs worldwide. Modifying the model parameters (e.g. thickening the late Eocene finite arc crustal thickness to 50 km) are inconsistent with results from geochemical modeling and would produce unrealistically high crustal growth rates. Crustal shortening is the main mechanism for thickening the crust in transpressional orogens, yet magmatic underplating may have been underestimated in the Andes, and grossly underestimated in non-transpressional orogens.