Continued Trenchward Procession of Upper Plate GPS Sites Following the 2012 Mw 7.6 Nicoya Earthquake

Abstract:

When studying subduction zone deformation one is often forced to consider a region significantly landward of the trench. The Nicoya Peninsula in Costa Rica presents a unique opportunity to obtain rich datasets from land in relatively close proximity to an active megathrust. A recent moment magnitude (Mw) 7.6 earthquake in September 2012 on this portion of the Middle America Trench affords an opportunity to constrain the ongoing postseismic deformation on the subduction interface between the Cocos and Caribbean plates. GPS campaigns occupying 22 sites were undertaken immediately following the earthquake in September-December 2012 and most recently in March 2015. Combined with data from a network of 17 continuous GPS in the region, we analyze the spatial and temporal changes in the postseismic velocity field. Another campaign is planned for 2017, in conjunction with our ongoing analysis of the continuous GPS network.

After 2.5 years, campaign GPS results indicate significant trenchward motion of at least 7 cm, relative to a fixed Caribbean plate, for all sites up to the volcanic chain. Maximum values of 22 cm are observed above and updip of the coseismic rupture zone. The trench-parallel component of the displacement field is small, with few deviations between sites. Together these observations are substantially more self-similar over a larger region than what was observed for the coseismic offset. This implies that there may be a low stress differential across the upper plate, suggesting that the subduction interface environment, including the mainshock and surrounding area, has remained relatively weak following the earthquake. By utilizing a dense and long-term geodetic network we will report on initial modeling that aims to characterize the evolution of afterslip. The effect of regional aftershocks, including an Mw 6.5 in October 2012, and viscoelastic mantle relaxation will be considered to establish the necessity of such effects in robustly accounting for postseismic deformation. We will also compare our afterslip results with observations of intereseismic locking and slow slip events for the region.