Geology of the Altiplano

The Geology of the Altiplano

For the past 200 Ma, subduction-related volcanism has defined the volcanic activity in the Central Andes. Around 26 Ma, shallowing of the subducting slab as a result of fast, nearly orthogonal convergence resulted in changes in convection and magma supply rates in the mantle wedge. These changes, accompanied by erosion of the lower part of the upper plate by stoping or delamination effectively increased the heat flux in the upper mantle and created a zone of weakness within the crust. Compressive failure of the weakened lithosphere later led to extensive crustal thickening from ~26-14 Ma (Isacks, 1988). Subsequently, the earliest magmatism in the Altiplana-Puna region occurred over a much broader area than the current Andean arc, consistent with a shallow angle of subduction. Steepening of the subducting slab to it’s present dip of 25 to 30° in the mid-Miocene (16-12 Ma) increased the volume of the mantle wedge and promoted melting of the hydrated lithosphere. The steepening of the dip was accompanied by delamination of the continental lithosphere (e.g. Kay et al. 1999; Allmendinger et al., 1997), evidence for lithospheric delamination includes high crustal elevations with lower crustal thicknesses suggesting the crust is supported by hotter more buoyant mantle that has risen into the lower crust and an apparent temporal relationship between the changes in subduction angle and an increased mantle power input that led to crustal melting and the APVC ignimbrite flare-up (Isacks, 1988; de Silva and Gosnold, 2007). 

Altiplano-Puna Volcanic Complex (APVC)

The APVC is a barren volcanic plateau located between 21° and 24° S with an average elevation of  ~4000 m. From 10-1 Ma an ignimbrite flare-up deposited at least 10,000 km3 (one estimate said 30,000 km?) of material over an area of about 70,000 km2 in the APVC (de Silva and Gosnold, 2007). Volcanic activity continued from 10 Ma to present, but was punctuated by distinct pulses of activity at 8, 6, and 4 Ma. Prior to 6 Ma activity was widespread across the plateau, however, after 6 Ma eruptions became more focused in the central APVC. The pulse at 4 Ma was the climactic stage of the APVC, and activity since 4 Ma has been comparatively minor, the largest eruptions being those from the Purico and Laguna Colorado centers at ~ 1 Ma (de Silva et al., 2006b).  

The ignimbrites from the APVC are all very similar in composition, and are typically of the large “monotonous intermediates” described by Hildreth (1981). They are high K dacites to rhyodacites with very volumetrically minor rhyolites. Andesites and basaltic-andesites are rare and occur as bands and inclusions in the pumices from some ignimbrites, and although volumetrically insignificant they may represent the thermal input into the upper-crustal Antiplano Puna Magma Body (APMB).

Prior to the ignimbrite flare-up, the APVC crust must have been thermally prepared. Delamination events discussed above led to a large increase in thermal input from the mantle. The increased thermal input along with with ongoing arc magmatism and crustal thickening thermally prepared APVC crust (elevated geotherms), resulting in extensive crustal melting. The impact of this increased thermal input was a period of intense generation, transport, and storage of thousands of km3 of magma in the mid crust. The elevated heat flux thermally softened the continental crust causing rapid elevation of the brittle-ductile transition zone. Increasing sizes of eruptions from 10-4 Ma indicate that progressively larger batches of magmas were being stored within the crust as the flare-up proceeded. Storage of increasingly large magma bodies may have been facilitated by changes in the mechanical properties of the wall rock surrounding the magma bodies due to the increased geotherms (i.e. wall rocks are thermally softened and more ductile). The elevated brittle-ductile transition zone coupled with regional tectonic stresses eventually resulted in major roof failures and the eruptions of the voluminous ignimbrites (e.g. de Silva et al., 2006a).


See Also