Continuously recording Global Positioning System stations near the 28 March 2005 rupture of the Sunda megathrust [moment magnitude (Mw) 8.7] show that the earthquake triggered aseismic frictional afterslip on the subduction megathrust, with a major fraction of this slip in the up-dip direction from the main rupture. Eleven months after the main shock, afterslip continues at rates several times the average interseismic rate, resulting in deformation equivalent to at least a M(w) 8.2 earthquake. In general, along-strike variations in frictional behavior appear to persist over multiple earthquake cycles. Aftershocks cluster along the boundary between the region of coseismic slip and the up-dip creeping zone. We observe that the cumulative number of aftershocks increases linearly with postseismic displacements; this finding suggests that the temporal evolution of aftershocks is governed by afterslip.

We determine coseismic and the first-month postseismic deformation associated with the Sumatra–Andaman earthquake of 26 December 2004 from near- field Global Positioning System (gps) surveys in northwestern Sumatra and along the Nicobar-Andaman islands, continuous and campaign gps measurements from Thailand and Malaysia, and in situ and remotely sensed observations of the vertical motion of coral reefs. The coseismic model shows that the Sunda subduction megathrust ruptured over a distance of about 1500 km and a width of less than 150 km, releasing a total moment of 6.7–7.0 × 1022 N m, equivalent to a magnitude Mw ∼9.15. The latitudinal distribution of released moment in our model has three distinct peaks at about 4° N, 7° N, and 9° N, which compares well to the latitudinal variations seen in the seismic inversion and of the analysis of radiated T waves. Our coseismic model is also consistent with interpretation of normal modes and with the amplitude of very-long-period surface waves. The tsunami predicted from this model fits relatively well the altimetric measurements made by the jason and topex satellites. Neither slow nor delayed slip is needed to explain the normal modes and the tsunami wave. The near-field geodetic data that encompass both coseismic deformation and up to 40 days of postseismic deformation require that slip must have continued on the plate interface after the 500-sec-long seismic rupture. The postseismic geodetic moment of about 2.4 × 1022 N m (Mw ∼8.8) is equal to about 30 ± 5% of the coseismic moment release. Evolution of postseismic deformation is consistent with rate-strengthening frictional afterslip. Online material: Summary of geodetic data used in this study.

[1] We document geodetic strain across the Nepal Himalaya using GPS times series from 30 stations in Nepal and southern Tibet, in addition to previously published campaign GPS points and leveling data and determine the pattern of interseismic coupling on the Main Himalayan Thrust fault (MHT). The noise on the daily GPS positions is modeled as a combination of white and colored noise, in order to infer secular velocities at the stations with consistent uncertainties. We then locate the pole of rotation of the Indian plate in the ITRF 2005 reference frame at longitude = À 1.34 AE 3.31 , latitude = 51.4 AE 0.3 with an angular velocity of W = 0.5029 AE 0.0072 /Myr. The pattern of coupling on the MHT is computed on a fault dipping 10 to the north and whose strike roughly follows the arcuate shape of the Himalaya. The model indicates that the MHT is locked from the surface to a distance of approximately 100 km down dip, corresponding to a depth of 15 to 20 km. In map view, the transition zone between the locked portion of the MHT and the portion which is creeping at the long term slip rate seems to be at the most a few tens of kilometers wide and coincides with the belt of midcrustal microseismicity underneath the Himalaya. According to a previous study based on thermokinematic modeling of thermochronological and thermobarometric data, this transition seems to happen in a zone where the temperature reaches 350 C. The convergence between India and South Tibet proceeds at a rate of 17.8 AE 0.5 mm/yr in central and eastern Nepal and 20.5 AE 1 mm/yr in western Nepal. The moment deficit due to locking of the MHT in the interseismic period accrues at a rate of 6.6 AE 0.4 Â 10 19 Nm/yr on the MHT underneath Nepal. For comparison, the moment released by the seismicity over the past 500 years, including 14 M W ≥ 7 earthquakes with moment magnitudes up to 8.5, amounts to only 0.9 Â 10 19 Nm/yr, indicating a large deficit of seismic slip over that period or very infrequent large slow slip events. No large slow slip event has been observed however over the 20 years covered by geodetic measurements in the Nepal Himalaya. We discuss the magnitude and return period of M > 8 earthquakes required to balance the long term slip budget on the MHT.