Control of Gravitational Potential Energy and associated stress field on crustal seismicity along the Andean margin

Lateral changes of lithospheric density structure and associated topography create spatial variations of Gravitational Potential Energy (GPE) that exert a primary control on the direction and magnitude of crustal stresses, the style of active faulting and, therefore, the location, spatial density and magnitude of crustal earthquakes. Zones of positive/negative GPE with respect to a stable region, should be characterized by an extensional/compressional stress regime, driving crustal deformation toward an ideal situation of spatially homogeneous GPE with no lateral gradients. Along active continental margins, these relationships can be altered by forces associated to subduction, namely the far-field tectonic forces due to plate convergence, elevated shear stresses along the interplate megathrust and basal drags driven by mantle wedge flow. Testing the role of GPE on crustal stresses and seismicity requires an adequate representation of the 3D density structure and a large dataset of stress field indicators and focal mechanism to allow a significant statistical comparison between model predictions and observations, both of which are commonly scarce.

In this contribution we will show results of a study performing this test along the Central and Southern Andean margin (5º-45ºS) that use a refined geophysically-constrained 3D density model, complemented by an analysis of Geoid anomalies, and a recently compiled dataset of several hundred stress tensors derived from Pliocene-to-Recent fault slip data and shallow earthquake focal mechanisms. These results show a strong first-order correlation between GPE anomalies and the large-scale stress field with positive/negative GPE correlating with normal/reverse faulting and near neutral GPE associated to strike-slip faulting. However, local misorientation of existing crustal faults with respect to this field causes stress rotations. First- and second-order partial derivatives of GPE are associated to the 2D stress tensor and compares well with the maximum horizontal stresses SHmax derived from the available data, confirming the main role of GPE on driving crustal deformation. This is further analyzed verifying a correlation between the spatial density of crustal seismic events and the magnitude of GPE gradients, which shed light about the level of stresses at crustal faults and the mechanism of their seismic activation. These results have important implications for understanding the forces driving crustal deformation and the controls on crustal seismicity in active orogenic systems.