It is usually assumed that short‐term (a few years) postseismic deformation around the rupture zone is caused by continuing slip (afterslip) along the fault interface, and viscoelastic stress relaxation is only responsible for long‐term deformation. In order to verify the validity of this assumption, the initial 1.5 months postseismic displacements following the 2015 Mw 8.3 Illapel earthquake are analyzed based on a multilayered structure model. We explore the possible mechanisms, including afterslip and viscoelastic relaxation, which might have contributed to the postseismic deformation, and aim to distinguish the contributing ratio of different postseismic processes. The results show that either the models of kinematic afterslip or viscoelastic stress relaxation individually cannot match the observed horizontal and vertical postseismic displacements satisfactorily. However, a combined model considering both afterslip and viscoelastic relaxation effects can reduce the data misfit significantly and is more physically reasonable. In the preferred combined model, the transient viscosities of the lower crust and upper mantle are 6×1017  Pas and 9×1017  Pas, respectively. The difference between the afterslip distribution of the pure afterslip model and that of the combined model indicates that previous models based on pure elastic assumption have substantially underestimated the afterslip updip of the rupture zone, and overestimated the afterslip downdip of the rupture zone. Therefore, the role of viscoelastic stress relaxation is indispensable in the study of transient postseismic deformation following a large earthquake, which contradicts the conventional concept about deformation mechanisms of early postseismic process.

You do not currently have access to this article.