Intermediate-depth earthquakes, slab stress state and upper mantle structure beneath the north central Andes

Flat slab subduction in Peru and the Central Andean Plateau of southern Peru, Bolivia and northwestern Argentina are the two intriguing geologic features along the western margin of South America. Flat slab subduction has often been causally linked to wide host of geological observations including the cessation of arc volcanism, inboard thick-skinned deformation of the overriding plate, ignimbrite volcanism, and the evolution of high plateaus. Yet several questions continue to exist on both the requirements for its formation and the consequences of its existence. For the purpose of this dissertation, I define "flat slab subduction" as subduction zones where the downgoing slab subducts normally (∼30° dip) to a depth of ∼100 km and then abruptly bends to travel horizontally for several hundred kilometers before resuming its descent. Understanding the mechanisms responsible for the formation of existing flat slabs will help us understand if they can be scaled up to match predictions for paleo-flat slabs. There are a number of proposed contributing factors to the formation of flat slab subduction, including ridge subduction, rapid overriding plate velocity and trenchward motion of a thick cratonic root. In this dissertation I investigate the role of Nazca ridge subduction on the formation of Peruvian flat slab in Chapter 1 and Chapter 2.;Previous studies of the geometry of the subducted Nazca plate in central and southern Peru have had to rely on primarily teleseismic data or local data collected from small seismic network. In Chapter 1 I determined the geometry of the subducting Nazca slab beneath central and southern Peru using data from the three recently deployed local seismic networks. I determined new contours of the slab geometry between 9° and 18°S using 508 relocated hypocenters and constraints from teleseismic surface wave tomography. This region offers a unique opportunity to study the link between ridge buoyancy and occurrence of the flat slab in general because of the unique ridge geometry relative to the convergence direction of the plates. My results show that the shallowest portion of the southern Peruvian flat slab is either at or just south of the subducted Nazca Ridge and events deepens to the north in the region of previously proposed flat or even shallower slab geometries. The fact that the shallowest portion of the Nazca plate is straddling the ridge and events deepen along strike and away from the ridge has important implications for the ridge buoyancy hypothesis. My relocated hypocenters also indicate an absence of seismicity along the projected location of the subducting Nazca Ridge, which is likely due to an absence of mantle hydrous phases beneath the overthickened crust of the Nazca Ridge. This provides an important insight for the genesis of intermediate depth seismicity.;The distribution of earthquake hypocenters as determined in Chapter 1 indicates strong along-strike variability in the slab geometry, from flat slab subduction north of 15°S, to uniform normal subduction south of 15°S. In Chapter 2 I obtained high quality focal mechanisms for intermediate depth events to investigate how the slab is deforming along strike. South of the Nazca Ridge, my results suggest uniform extension in the slab down dip of the ridge. Down dip tension is consistent with a highly deformed slab, but no tearing between the normally dipping plate and the flat slab along the Nazca Ridge. North of the Nazca Ridge, the T-axes are largely ridge-parallel in map view, but with a distinct downward dip that is not parallel to the slab. These steeply dipping T-axes differ from the expected stress pattern for a fully supported flat slab and indicate that the flat slab north of the ridge may not stable.;South of the Peruvian flat slab, the Central Andean Plateau (i.e. Altiplano-Puna plateau) is the second largest tectonically active orogen along the western margin of South America. This plateau has influenced both local and far field lithospheric deformation, global sediment flux, atmospheric circulation and climate since the early Miocene. Significant geologic and geophysical efforts have been made to constrain the tectonomorphic evolution of the central Andean plateau, yet the role of surface and deep lithospheric processes in the evolution of the plateau is unclear. Existing theories predict two contrasting models of rapid and recent versus slow and steady uplift for the temporal evolution of the Central Andean Plateau. One possible discriminating factor between these two theories is seismic evidence for the presence or absence of mantle lithosphere. In Chapter 3 I investigate the current state of lithospheric structure below the northern Altiplano, northernmost portion of the Central Andean Plateau, in southern Peru and northern Bolivia. My results indicate an absence of a high velocity lower crust beneath the northern Altiplano, suggesting a weak lower crust of felsic composition or the loss of a high velocity mafic lower crust due to delamination. The upper mantle under the northern Altiplano is heterogeneous, consistent with piecemeal delamination. My tomography results for the lower crust and upper mantle beneath the northern Altiplano are in better agreement with the slow and steady uplift model.