Decoding Upper-Plate Aftershocks: The Critical Role of Pore-Pressure Diffusion following the 2014 Iquique Earthquake

After large earthquakes, aftershocks are observed globally as a time-dependent phenomenon. In subduction zones, aftershocks occurring in the upper plate are particularly hazardous, as they often take place near densely populated areas, increasing the risk to structures already weakened by the mainshock. The number of aftershocks typically decreases over time, following a pattern described by the empirical Omori-Utsu law. Despite this well-documented behavior, the physical mechanisms driving this decay remain uncertain. While coseismic static stress transfer cannot explain the non-linear time dependence of aftershocks, transient postseismic processes such as afterslip and viscoelastic relaxation have been proposed as possible mechanisms. Alternatively, considering the temporal decay of aftershock sequences and the similar behavior observed in induced seismicity caused by wastewater injection, we explore the hypothesis that pore-pressure diffusion plays a key role in controlling the spatial and temporal distribution of natural earthquake aftershocks.

In this study, we investigate the 2014 Mw 8.2 Iquique event to test our hypothesis, using an approach that integrates geodetic and seismological data, as well as geological, frictional, rheological, and hydraulic constraints. Using a 4D (space and time) modeling approach considering realistic rock material properties, we first reproduce the 3D postseismic deformation time series observed by continuous GNSS stations. We then disaggregate the individual contributions of the three dominant postseismic processes, i.e., afterslip, viscoelastic, and poroelastic relaxation, to the deformation signal. In particular, poroelastic deformation substantially affects the observed vertical geodetic signal in the near field. We then compute and analyze the spatiotemporal stress changes produced by the individual postseismic processes using the Coulomb Failure Stress (CFS) parameter. By comparing these CFS changes to the distribution of upper-plate aftershocks, we find that stress changes produced by pore-pressure changes best correlate in space with increased upper-plate aftershock activity. Furthermore, increased pore pressure reduces the effective fault normal stresses independently of the fault orientation and consequently triggers all faulting styles. This explains the higher diversity of faulting styles observed in upper-plate aftershocks. Finally, we find a very strong temporal correlation (>0.98) between the exponential increase of the cumulative number of upper-plate aftershocks and pore-pressure changes. This finding suggests that the unclear physical basis for Omori-type aftershock decay may relate to the hydraulic properties (e.g., rock permeability and porosity) of the upper plate. Thus, our work offers a deeper understanding of the hydro-mechanical behavior of the upper plate during large earthquakes and may open new avenues for physics-based aftershock forecasting.