Variations in Forearc Stress and Changes in Principle Stress Orientations Caused by the 2004–2005 Megathrust Earthquakes in Sumatra, Indonesia

Coseismic changes in principal stress orientation in the northern Sumatra subduction zone due to two giant megathrust earthquakes there in 2004 and 2005 are estimated to investigate the in-situ stress. The two megathrust earthquakes, the 2004 Sumatra-Andaman and the 2005 Nias-Simeulue events, are both among the 11 largest earthquakes ever recorded. Previous studies have shown that these giant earthquakes perturbed the stress field in the Sumatra subduction zone enough to alter the principal stress directions there, and here we investigate whether these changes can be used to better understand spatial variations in stress along the subduction zone. We used 330 previously published focal mechanisms to estimate pre- and post-mainshock principal stress orientations in 3 outer forearc segments and assessed whether orientation differences were resolved and what they imply about the pre- and post-mainshock stress fields. Our results agree with previous studies in establishing that coseismic changes in stress orientation in the forearc are resolvable, and consistent with a low level of stress in the outer Sumatran forearc before the earthquake, with almost all the shear stress on the megathrust relieved in the 2004 and 2005 earthquakes. In this study, we reveal that both the stress orientations and coseismic changes in them exhibit along-strike variations, with a decrease in both the pre-mainshock stress and stress drop found in the rupture area of 2005 relative to that of the 2004 earthquake. The forearc segment between the 2004 and 2005 rupture areas, which coincides with a well-known megathrust rupture barrier beneath the island of Simeulue is observed to have a characteristic signature, with lower shear stress relative to the pre-mainshock stress field and higher shear stress relative to the post-mainshock stress field in the adjacent segments.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.