How Does Trench Coupling Lead to Mountain Building in the Sub-Andes?

The Andes are commonly believed to have resulted from subduction-related crustal shortening. Recent GPS data show ~30-40 mm/yr crustal shortening uniformly distributed across the central Andes, contrasting to geological observations of slower shortening (~10-15 mm/yr) that is concentrated in the sub-Andean fold and thrust belt. Previous studies have interpreted GPS-mesured crustal shortening to be mainly transient that will be recovered during future earthquakes, consistent with frequent trench earthquakes and the associated coseismic and postseismic crustal rebound. So how does the cyclic crustal shortening and rebound lead to long- term mountain building in the sub-Andes? To address this question, we have developed a 2D viscoelastic-plastic finite element model to investigate the linkage between short- and long-term strain partitioning across the central Andes. We simulate strain evolution during subduction earthquake cycles, and explore the effects of major model parameters and some of the inferred geological processes. Our results show that subduction alone is inadequate for building the Andes, because much of the interseismic crustal shortening in the upper plate may be recovered by coseismic slip and postseismic relaxation. We find that plastic deformation in the sub-Andes, in the form of sliding on the detachment faults or plastic failure of the sedimentary cover, is necessary to lead to the observed crustal shortening. Furthermore, our model suggest that numerous factors including topographic load of the Altiplano plateau, stronger trench coupling, faster western drift of the overriding South American plate, and the inferred delamination of mantle lithosphere beneath the eastern Cordillera may have contributed to the localized and apparently accelerated mountain building in the sub-Andes during the past ~10 Ma.