The Time-Space Distribution of Cenozoic Volcanism in the South-Central Andes: a New Data Compilation and Some Tectonic Implications

The coincidence of late Paleogene to Neogene shortening and crustal thickening with vigorous volcanic activity in the central Andes has long invited speculation about a causal relationship between magmatism and deformation. In aid of understanding this and related issues, we present here a new compilation of radiometric ages, geographic location and dominant rock type for about 1450 Cenozoic volcanic and subvolcanic centers in the southcentral Andes (14–28° S). This paper describes variations in the timespace distribution of volcanism from 65 to 0 Ma, with emphasis on the post-30 Ma period where Andean-style shortening deformation and volcanism were most intense. The central Andes are unusual for the abundance of felsic ignimbrites and their distribution is shown separately from the intermediate to mafic volcanic centers which are here termed the “arc association”. Overall, the time-space patterns of volcanic activity for the ignimbrite and the arc association are similar but ignimbrite distribution is more patchy and more closely associated spatially with the plateau region.The distribution of volcanic activity as a function of longitude and age, as well as cumulative frequency curves of volcanic centers as a function of age reveal major differences in arc productivity, i.e., number vs. age of volcanic centers, from north to south along the arc. Eocene and early-mid Oligocene activity was confined to a narrow belt in the Precordillera. Post-30 Ma activity was shifted to the east and spread over a much broader area than earlier arcs, probably due to a shallower subduction angle. This phase of volcanism began at about the same time from north to south (ca. 25 Ma) but the peak activity shifted progressively southward with time. Cumulative frequency curves demonstrate that 50% of volcanic output north of 20° S accumulated by 16 Ma, whereas this level was reached for 20–23°, 23–26° and 26–28° S segments at about 12, 10 and 8 Ma, respectively. Plate reconstructions place the subducted part of the Juan Fernández Ridge beneath the arc at these latitudes between about 25 and 5 Ma, but age-frequency diagrams of volcanism show no evidence that ridge subduction influenced arc productivity.The spatial distribution of volcanism shows some influence by crustal structures on a local scale (tens of km), most notably the preferential clustering of volcanic centers at intersections of the frontal arc with NW-SE-trending lineament zones. However, on a regional scale the time-space distribution of volcanic centers and the distribution of active shortening domains in the central Andes varied independently. The evidence does not support the concept that Andean crustal thickening and plateau formation were preconditioned by thermal weakening of the crust. A comparison of volcanic output vs. shortening rates for latitude 19–22° S confirms that the onset of intense deformation in the Oligocene preceded that of volcanism by about 10 Ma, and the increase in volcanic activity at about 20–16 Ma has no expression in shortening rates. After plateau formation, however, beginning at about 10 Ma, both shortening rate and volcanic output increased together and reached their highest levels. This period experienced extremely large-volume ignimbrite eruptions from the Altiplano-Puna volcanic complex. The ignimbrite magmas represent an episode of widespread crustal melting, and it is likely that the rise in shortening rates reflects meltenhanced weakening of the crust.Variations in the location and width of the CVZ arc respond to changes in slab dip, but the complex distribution of volcanic centers in time and space shown in this study belies a simple relationship. A condition that must be met for correlation between surface volcanism and slab dip is a near-vertical ascent of magmas, both in the mantle wedge and through the crust. We conclude that in the Neogene arc of the central Andes, vertical ascent of magmas is disturbed by effects of crustal heterogeneity, intense deformation, lithospheric thickening and partial delamination, and crustal melting.