Large earthquakes produce crustal deformation that can be quantified by geodetic measurements, allowing for the determination of the slip distribution on the fault. We used data from Global Positioning System (GPS) networks in Central Chile to infer the static deformation and the kinematics of the 2010 moment magnitude (M(w)) 8.8 Maule megathrust earthquake. From elastic modeling, we found a total rupture length of ~500 kilometers where slip (up to 15 meters) concentrated on two main asperities situated on both sides of the epicenter. We found that rupture reached shallow depths, probably extending up to the trench. Resolvable afterslip occurred in regions of low coseismic slip. The low-frequency hypocenter is relocated 40 kilometers southwest of initial estimates. Rupture propagated bilaterally at about 3.1 kilometers per second, with possible but not fully resolved velocity variations.

Abstract. It is generally accepted that the Aden ridge has propagated westward from -58øE to the western tip of the Gulf of Aden/Tadjoura, at the edge of Afar. Here, we use new tectonic and geochronological data to examine the geometry and kinematics of deformation related to the penetration of that ridge on dry land in the Republic of Djibouti. We show that it veers northward, forming a narrow zone of dense faulting along the northeastern edge of the Afar depression. The zone includes two volcanic rifts (Asal-Ghoubbet and Manda Inakir), connected to one another and to the submarine part of the ridge by transfer zones. Both rifts are composite, divided into two or three disconnected, parallel, NW-SE striking subrifts, all of which appear to have propagated northwestward. In Asal-Ghoubbet as in Manda Inakir, the subrifts appear to have formed in succession, through north directed jumps from subrifts more farther south. At present, the northernmost subrifts (Manda and Dirko Koma) of the Manda Inakir rift, form the current tip of the northward propagating Arabia-Somalia plate boundary in Afar. We account for most observations by a mechanical model similar to that previously inferred for the Gulf of Aden, in which propagation is governed by the intensity and direction of the minimum horizontal principal stress, (•3. We interpret the northward propagation on land, almost orthogonal to that in the gulf, to be related to necking of the Central Afar lithosphere where it is thinnest. Such necking may be a consequence of differential magmatic thickening, greater in the center of the Afar depression where the Ethiopian hot spot enhanced profuse basaltic effusion and underplating than along the edges of the depression. The model explains why the Aden ridge foregoes its WSW propagation direction, constant from -58øE to Asal-Ghoubbet. At a smaller scale, individual rifts and subrifts keep opening perpendicular to the Arabia-Somalia (or Danakil-Somalia) motion vector and propagate northwestward. Concurrently, such lithospheric cracks are forced to jump northward, such that the plate boundary remains inside the regional N-S necking zone. Changes of obliquity between the directions of overall and local propagation may account for different segmentation patterns, a small angle promoting long, en 6chelon subrifts, and a high-angle, smaller, nested, "subrifts within subrifts." The propagation mechanism is thus similar, whether in oceanic or continental lithosphere, the principal change being the overall propagation path, here governed by thickness changes rather than by the geometry in map view as previously inferred for the rest of the Aden ridge. Finally, because the same mechanism has led rifting along the Red Sea to propagate southward and jump to the western edge of Afar, the Arabia-Somalia and ArabiaNubia plate boundaries tips have missed each other and keep overlapping further, leading to strain transfer by large-scale bookshelf faulting.

Abstract.Between 1990 tectonic observations shows that these two earthquakes occurred on low-angle (_< 35 ø) north dipping normal faults located between 4.5 and 10 km depth in the inner part of the rift. Assuming that the deformation is concentrated in relatively narrow deforming zones, we use a simple model of a dislocation in an elastic half-space to study the implication of the localization. Using the geometry of the known seismogenic faults, our observations imply continuous aseismic deformation in the uppermost crust of the inner rift. This model predicts geodetic strain rates close to seismic strain rates in opposition to previous estimates. This is because our model takes into account the activity on low-angle normal faults in the inner rift and an effective seismogenic layer of 6-7 kin, about half that usually assumed.

Previous Caribbean GPS studies have shown that the rigid interior of the Caribbean plate is moving east‐northeastward (070°) at a rate of 18–20 ± 3 mm/yr relative to North America. This direction implies maximum oblique convergence between the island of Hispaniola on the Caribbean plate and the 22–27‐km‐thick crust of the Bahama carbonate platform on the adjacent North America plate. We present a tectonic interpretation of a 15‐site GPS network which spans the Hispaniola‐Bahama oblique collision zone and includes stable plate interior sites on both the North America and Caribbean plates. Measurements span the time period of 1994–1999. In a North America reference frame, GPS velocities in Puerto Rico, St. Croix, and the Lesser Antilles indicate that these areas move as a single block in an east‐northeast direction (070°) at a rate of 19–20 mm/yr consistent with the movement of the rigid interior of the Caribbean plate. GPS velocities at six sites in central and eastern Hispaniola (Dominican Republic) show drastically different behavior with more eastwardly strikes (080°) and much slower rates (4–17 mm/yr) than areas of the stable Caribbean plate. The boundary between the relatively slower‐moving Hispaniola collisional zone and the relatively faster‐moving, uncollided Puerto Rico‐Virgin Islands area is the Mona Passage where late Neogene rifting occurs in a broad zone. Elastic modeling favors strain partitioning with oblique slip on the outer, low‐angle submarine thrust faults (North Hispaniola, Muertos) and strike slip on the inner, subvertical subaerial, strike‐slip faults (Septentrional, Enriquillo).

SUMMARY The different phases of the earthquake cycle can produce measurable deformation of the Earth's surface. This work is aimed at describing the evolution of that deformation in space and time, as well as the distribution of causal slip on the fault at depth. We have applied GPS and synthetic aperture radar (SAR) interferometry (InSAR) techniques to northern Chile, where fast plate convergence rates are associated with large subduction earthquakes and extensive crustal deformation. The region of northern Chile between 18°S and 23°S is one of the most important seismic gaps in the world, with no rupture having occurred since 1877. In 1995, the Mw= 8.1 Antofagasta earthquake ruptured the subduction interface over a length of 180 km in the region immediately to the south of this 450 km long gap. The coseismic deformation associated with this event has been documented previously. Here we use GPS position time‐series for 40 benchmarks (measured between 1996 and 2000) and ERS SAR interferograms (for the interval between 1995 and 1999) to map both the post‐seismic deformation following the 1995 event and the ongoing interseismic deformation in the adjacent gap region. In the seismic gap, the interseismic velocities of 20–30 mm yr−1 to the east with respect to South America are mapped. Both the GPS and the InSAR measurements can be modelled with 100 per cent coupling of the thrust interface of the subduction to a depth of 35 km, with a transition zone extending down to 55 km depth. The slip rate in that zone increases linearly from zero to the plate convergence rate. South of the gap, the interferometric map shows interseismic deformation superimposed with deformation following the 1995 earthquake and covering the same area as the coseismic deformation. Some 40 per cent of this deformation is related to seismic activity in the 3.3 yr following the 1995 event, in particular slip during a Mw= 7.1 earthquake in 1998. However, most of the signal (60 per cent) corresponds to post‐seismic deformation resulting from widespread aseismic slip in the subduction interface. The afterslip appears to have occurred down‐dip in the transition zone at 35–55 km depth and to have propagated laterally northwards at 25–45 km depth under the Mejillones Peninsula, which is a prominent geomorphological feature at the boundary between the 1877 and 1995 rupture zones. We propose a simple slip model for the seismic cycle associated with the Antofagasta earthquake, where the transition zone alternates between aseismic shear and seismic slip.

11The Concepción-Constitución area [35-37°S]

[1] During the last 10 years, a network of about 30 GPS sites was measured in Djibouti, East Africa. Additional points were also measured in Yemen, Oman, Ethiopia, Iran, and on La Réunion island. Merged with data from the available International GPS Service permanent stations scattered on the different plates in the area (Eurasia, Anatolia, Africa, Arabia, Somalia), this unique data set provides new insight on the current deformation in the Africa-Somalia-Arabia triple junction area and on the Arabian plate motion. Here we show that coherent motions of points in Yemen, Bahrain, Oman, and Iran allow us to estimate a geodetically constrained angular velocity for the Arabian plate (52.59°N, 15.74°W, 0.461°/Myr in ITRF2000). This result differs significantly from earlier determinations and is based upon our vectors in Yemen. They provide new additional data and better geometry for angular velocity determination. Combined with the African and Somalian motions, this new angular velocity results in predicted spreading rates in the Red Sea and the Gulf of Aden which are 15-20% lower than those measured from oceanic magnetic anomalies and thus averaged over the last 3 Myr. With respect to Eurasia, the geodetic motion of Arabia is also about 30% slower than predicted by NUVEL-1A. On the basis of the kinematic results presented here and on other evidence for a similar slower geodetic rate of the Indian plate, we suggest that the whole collision zone between Africa, Arabia, India on one hand and Eurasia on the other hand has slowed down in the last 3 Myr.

[1] We analyzed aftershocks and postseismic deformation recorded by the continuous GPS station AREQ following the M w = 8.4, 23 June 2001 Peru earthquake. This station moved by 50 cm trenchward, in a N235°E direction during the coseismic phase, and continued to move in the same direction for an additional 15 cm over the next 2 years. We compare observations with the prediction of a simple one-dimensional (1-D) system of springs, sliders, and dashpot loaded by a constant force, meant to simulate stress transfer during the seismic cycle. The model incorporates a seismogenic fault zone, obeying rate-weakening friction, a zone of deep afterslip, the brittle creep fault zone (BCFZ) obeying rate-strengthening friction, and a zone of viscous flow at depth, the ductile fault zone (DFZ). This simple model captures the main features of the temporal evolution of seismicity and deformation. Our results imply that crustal strain associated with stress accumulation during the interseismic period is probably not stationary over most of the interseismic period. The BCFZ appears to control the early postseismic response (afterslip and aftershocks), although an immediate increase, by a factor of about 1.77, of ductile shear rate is required, placing constraints on the effective viscosity of the DFZ. Following a large subduction earthquake, displacement of inland sites is trenchward in the early phase of the seismic cycle and reverse to landward after a time t i for which an analytical expression is given. This study adds support to the view that the decay rate of aftershocks may be controlled by reloading due to deep afterslip. Given the ratio of preseismic to postseismic viscous creep, we deduce that frictional stresses along the subduction interface account for probably 70% of the force transmitted along the plate interface.