Seismic studies in the southern Puna plateau and the Peruvian Andes

I present three studies in two regions, both within the Central Andes. In both regions it is clear that there are significant variations in the subduction geometry. I have used surface wave tomography to investigate the shear wave velocity structure beneath the southern Puna plateau and found evidence of widespread melting of the crust beneath the high elevations which correlates with a gap in intermediate depth seismicity and the recent eruptions of ignimbrite complexes. All of these observations can be explained by the delamination of the lithospheric mantle beneath it. I measured Rayleigh wave phase velocities as a function of frequency and inverted then to obtain shear wave velocities as a function of depth. The results show a high velocity body sitting above the subducting Nazca plate beneath the northern edge of the Cerro Galan ignimbrite. This high velocity body is interpreted to be the delaminated piece of lithosphere that detached and sank leading to a localized upwelling of asthenosphere that, in turn, caused widespread crustal melting leading to the eruption of the most recent ignimbrite complexes. Furthermore it is possible that this upwelling also thermally eroded the slab in this region. It is apparent that there is a significant slab gap or hole where there are very few intermediate depth earthquakes.;In addition, I have used shear wave splitting analysis and shear wave splitting tomography in the southern Puna plateau to investigate the patterns of seismic anisotropy and mantle flow. The results show very complex shear wave splitting and seismic anisotropy patterns throughout the southern Puna plateau. The observations suggest that different mechanisms are driving the mantle flow from region to region. The subslab mantle outside the region where delamination took place is mostly driven by slab roll back and small degree of coupling between the subducting slab and the mantle below it. In the region apparently dominated by delamination, the subslab mantle outlines a rotation pattern of a-axis and small strength of anisotropy, suggesting that a slab gap or heated slab is regionally driving the mantle flow. In the mantle wedge, there is evidence for vertical a-axis suggesting vertical flow. This is observed beneath the main ignimbrite complexes and above the delaminated block imaged in the first study. Anisotropy in the western edge of the region above the slab is driven by compression of the South American lithosphere, which leads to trench parallel a-axis orientation.;Finally, I have studied the attenuation of high frequency seismic waves in Southern Peru using the two-station and reverse two station methods. I have focused on the largest regional phase Lg which is a wave that is entirely guided in the crust and travels mostly as an S-wave. This study allowed me to investigate the correlation of attenuation measurements with topography and regions of high heat flow. Furthermore, this study is important as it can be used to distinguish regions with higher seismic hazard from distance earthquakes. Low attenuation, or high quality factor (Q), usually leads to high stronger ground shaking even at larger distances from the epicenter. Our attenuation model should determine whether the attenuation is intrinsic or cause by scattering. The western cordillera in northern Peru is characterized by high attenuation. I have also found that most regions with high attenuation (low LgQ) in southern Peru correlated very well with location of volcanoes. However, we have also found regions with high attenuation where there is very little active volcanism. We believe the high level of attenuation is associated with scattering attenuation caused by a thick crustal root. The Altiplano plateau, where Lake Titicaca is located, shows intermediate values of attenuation. The city of Pisco shows the highest values of Q in Peru (∼1500) associated with a high degree of site effect which makes this region potentially more susceptible to strong ground motion from regional earthquakes. Most of southern Peru tends to exhibit high intrinsic attenuation but scattering attenuation is also observed to dominate in regions where the topography changes drastically over small distances such as the northern edge of the Altiplano plateau.