Neogene magmatic history and geodynamic evolution of the southern Andean Puna plateau.

KAY, S.M.; COIRA, B. Y C., MPODOZIS

Potsdam, Germany.

Simposio; 19 th Colloquium on Latin American Geociences.; 2005

IGCP 508 Project-IUGS-UNESCO

Over 400 chemical analyses of middle Miocene to Holocene arc and backarc magmatic rocks between 25°S and 28°S provide clues to the mantle and crustal lithospheric evolution of the southern Puna plateau. Temporal changes in magmatic distribution and chemistry are consistent with a transient and subdued shallowing of the downgoing Nazca plate as the Juan Fernandez ridge subducted beneath the region at about 12 to 7 Ma.  No magmatic gap was produced.  This shallowing set the stage for Pliocene lower crustal and mantle lithospheric removal (‘delamination’) by thinning the continental lithosphere and likely accelerating eclogite formation in a thickened mafic lower crust.  Contemporaneous processes include redistribution of crustal and mantle volume by lateral flow, eastward displacement of the frontal arc, and loss of crustal mass by forearc subduction erosion.  The eastward expansion of ~ 15 to 8 Ma andesitic/dacitic stratovolcanoes and dome complexes with arc-like trace element signatures into the backarc is consistent with a subducted component above a moderately shallow slab.  By 6.6 Ma, mafic lavas in the region had arc-like La/Ta ratios of 36-55 with ratios reaching 45-78 in ~ 5-4.3 Ma glassy andesites and > 64 in glassy dacites.  Mafic magmas < 3 Ma in the central region revert to more subdued to intraplate-like ratios (20-35) whereas those to the north and south remain arc-like.  Voluminous Cerro Galán region ignimbrites to the east have La/Ta ratios of 27-42 at ~ 6.6-4 Ma and 14-28 after ~ 2.4 Ma.  Parallel changes in mafic and silicic magmas are consistent with modification of the lower crust by interaction with mantle-derived magmas.  Less arc-like signatures in central southern Puna magmas after 3 Ma fit with delamination of lower crust and mantle lithosphere as the slab steepened beneath the region Based on the literature for same region, a comparison of 2-D crustal shortening estimates (< 100 km) from field data with estimates of crustal shortening (>250 km) from topographic-based crustal cross-sectional area shows a large mismatch.  These estimates suggest that the crust, whose thickness is not well known, could not have thickened by backarc crustal shortening at this latitude alone. Analyses of trace element data show that some ca 5-4 Ma dacites equilibrated with residual mafic eclogitic crust at high pressure before erupting.  One way to reconcile these observations is for the crust to have thickened in response to deep crustal flow from the north and the south. This flow is consistent with a crust warmed by magmatic intrusions and could be supported by isotopic evidence for a more radiogenic crustal component in  post 7 Ma ignimbrites.  Maximum flow likely occurred near 9 to 6 Ma in association with backarc thrusting and rapid uplift during a climax of backarc ignimbrite activity to the north, uplift of the Pampean ranges to the east and shallowing of the subduction zone to the south. Other considerations are the consequences of up to 50 km of eastward displacement of the arc front in the last 20 Ma.