S U M M A R YRecent dense deployments of portable digital seismographs have provided excellent control on earthquakes beneath the central North Island of New Zealand. Here we use a subset of the best-recorded earthquakes in an inversion for the 3-D Vp and Vp/Vs structure. The data set includes 39 123 P observations and 18 331 S observations from 1239 earthquakes and nine explosions. The subducted plate is imaged as a high Vp, low Vp/Vs feature. Vp within the mantle of the subducted slab is almost always >8.5 km s −1 , which requires the ca. 120 Myr slab to be unusually cold. The low Vp/Vs within the subducted plate closely parallels the lower plane of the dipping seismic zone. It most likely indicates fluid resulting from dehydration of serpentine in the slab mantle, and the earthquakes themselves are likely to be promoted by dehydration embrittlement. We identify a region with Vp < 8.0 km s −1 which coincides with the upper plane of the dipping seismic zone and extends to ca. 65 km depth with the subducted Hikurangi Plateau, which is about 17 km thick prior to subduction. The mantle wedge is generally imaged as a low Vp, high Vp/Vs feature. However, there are significant changes evident in the wedge along the strike of the subduction zone. The region where Vp is lowest (7.4 km s −1 ) and Vp/Vs is highest (1.87) occurs at 65 km depth, immediately west of the Taupo caldera. This region is best interpreted as a significant volume of partial melt, produced by the reaction of fluid released by dehydration of the subducted plate with the convecting mantle wedge. The region with lowest Vp, while paralleling the underlying dipping seismic zone, is located about 30 km from the upper surface of the zone. Material with Vp > 8.0 km s −1 directly above the dipping seismic zone can be interpreted as sinking, entrained with the motion of the subducted slab and forming a viscous blanket that insulates the slab from the high-temperature mantle wedge. Material in the overlying low Vp region can be interpreted as rising within a return flow within the wedge. The volcanic front appears to be controlled by where this dipping low Vp region meets the base of the crust. The thickness of the backarc crust also shows significant variation along strike. In the central Taupo Volcanic Zone (TVZ) the crust is ca. 35 km thick, while southwest of Mt Ruapehu the crust thickens by ca. 10 km. There is no significant low Vp zone in the mantle wedge in this southwestern region, suggesting that this thicker crust has choked off mantle return flow. The seismic tomography results, when combined with constraints on mantle flow from previous shear-wave splitting results, provide a plausible model for both the distribution of volcanism in the central North Island, and the exceptional magmatic productivity of the central TVZ.

The rifting of continents and evolution of ocean basins is a fundamental component of plate tectonics, yet the process of continental break-up remains controversial. Plate driving forces have been estimated to be as much as an order of magnitude smaller than those required to rupture thick continental lithosphere 1,2 . However, Buck 1 has proposed that lithospheric heating by mantle upwelling and related magma production could promote lithospheric rupture at much lower stresses. Such models of mechanical versus magma-assisted extension can be tested, because they predict different temporal and spatial patterns of crustal and upper-mantle structure. Changes in plate deformation produce strain-enhanced crystal alignment and increased melt production within the upper mantle, both of which can cause seismic anisotropy 3 . The Northern Ethiopian Rift is an ideal place to test break-up models because it formed in cratonic lithosphere with minor far-field plate stresses 4,5 . Here we present evidence of seismic anisotropy in the upper mantle of this rift zone using observations of shear-wave splitting. Our observations, together with recent geological data, indicate a strong component of melt-induced anisotropy with only minor crustal stretching, supporting the magma-assisted rifting model in this area of initially cold, thick continental lithosphere.The data we analysed were collected as part of the EAGLE project (Ethiopian Afar Geophysical Lithospheric Experiment), an international multi-institutional experiment designed to investigate rifting processes in Ethiopia 6 . The Miocene-Recent Ethiopian Rift (Fig. 1) constitutes the northern part of the East African Rift system and forms one arm of a triple junction that formed on or near a mantle plume. Our study region is transitional between continental and incipient oceanic, with strain localized to ,20-km-wide zones of dyking, faulting and volcanism 6,7 . It is an ideal place to study magmatism and plate rupture, because up to 25% of the crust is extruded lava or intrusive magma 7,8 and mantle lithosphere is thin (,50 km) beneath the rift valley 9 . Seismic data were acquired in three phases of the EAGLE project 6 , two of which were designed to record passive seismicity. In phase I, 29 broad-band seismometers were deployed for 16 months with a nominal station spacing of 40 km and covering a 250 km £ 350 km region centred on the transitional part of the rift (Fig. 1). In phase II, a further 50 instruments were deployed for three months in a tighter array (nominal station spacing of 10 km) in the rift valley. Our study of mantle anisotropy is based on evidence of shear-wave splitting in the teleseismic phases SKS, SKKS and PKS recorded by these two arrays. With the longer duration array, 15 events produced usable splitting results, and with the shorter duration rift-valley array, three events produced usable results (list of events given in Supplementary Information).Shear-wave splitting analysis of the seismic phases SKS, SKKS and PKS is now a standard tool for studyi...

[1] The Miocene-Recent East African Rift in Ethiopia subaerially exposes the transitional stage of rifting within a young continental flood basalt province. As such, it is an ideal study locale for continental breakup processes and hot spot tectonism. We combine teleseismic traveltime data from 108 seismic stations deployed during two spatially and temporally overlapping broadband networks to present detailed tomographic images of upper mantle P and S wave seismic velocity structure beneath Ethiopia. Tomographic images reveal a $500 km wide low P and S wave velocity zone at 75 to !400 km depth in the upper mantle that extends from close to the eastern edge of the Main Ethiopian Rift (MER) westward beneath the uplifted and flood basalt-capped NW Plateau. We interpret this broad low-velocity region (LVR) as the upper mantle continuation of the African Superplume. Within the broad LVR, zones of particularly low velocity are observed with absolute delay times (dt P $ 4 s) that indicate the mantle beneath this region is amongst the slowest worldwide. We interpret these low velocities as evidence for partial melt beneath the MER and adjacent NW Plateau. Surprisingly, the lowest-velocity region is not beneath Afar but beneath the central part of the study area at $9°N, 39°E. Whether this intense low-velocity zone is the result of focused mantle upwelling and/or enhanced decompressional melting at this latitude is unclear. The MER is located toward the eastern edge of the broad low-velocity structure, not above its center. This observation, along with strong correlations between low-velocity zones and lithospheric structures, suggests that preexisting structural trends and Miocene-to-Recent rift tectonics play an important role in melt migration at the base of the lithosphere in this magmatic rift zone.

The volcanically active Main Ethiopian rift (MER) marks the transition from continental rifting in the East African rift to incipient seafloor spreading in Afar. We use new seismicity data to investigate the distribution of strain and its relationship with magmatism immediately prior to continental breakup. From October 2001 to January 2003, seismicity was recorded by up to 179 broadband instruments that covered a 250 km × 350 km area. A total of 1957 earthquakes were located within the network, a selection of which was used for accurate location with a three‐dimensional velocity model and focal mechanism determination. Border faults are inactive except for a cluster of seismicity at the structurally complex intersection of the MER and the older Red Sea rift, where the Red Sea rift flank is downwarped into the younger MER. Earthquakes are localized to ∼20‐km‐wide, right‐stepping en echelon zones of Quaternary magmatism and faulting, which are underlain by mafic intrusions that rise to 8–10 km subsurface. Seismicity in these “magmatic segments” is characterized by low‐magnitude swarms coincident with Quaternary faults, fissures, and chains of eruptive centers. All but three focal mechanisms show normal dip‐slip motion; the minimum compressive stress is N103°E, perpendicular to Quaternary faults and aligned volcanic cones. The earthquake catalogue is complete above ML 2.1, and the estimated b value is 1.13 ± 0.05. The seismogenic zone lies above the 20‐km‐wide intrusion zones; intrusion may trigger faulting in the upper crust. New and existing data indicate that during continental breakup, intrusion of magma beneath ∼20‐km‐wide magmatic segments accommodates the majority of strain and controls the locus of seismicity and faulting in the upper crust.

[1] The Afar depression is an ideal locale to study the role of extension and magmatism as rifting progresses to seafloor spreading. Here we present receiver function results from new and legacy experiments. Crustal thickness ranges from ∼45 km beneath the highlands to ∼16 km beneath an incipient oceanic spreading center in northern Afar. The crust beneath Afar has a thickness of 20-26 km outside the currently active rift segments and thins northward. It is bounded by thick crust beneath the highlands of the western plateau (∼40 km) and southeastern plateau (∼35 km). The western plateau shows V P /V S ranging between 1.7-1.9, suggesting a mafic altered crust, likely associated with Cenozoic flood basalts, or current magmatism. The southeastern plateau shows V P /V S more typical of silicic continental crust (∼1.78). For crustal thicknesses <26 km, high V P /V S (>2.0) can only be explained by significant amounts of magmatic intrusions in the lower crust. This suggests that melt emplacement plays an important role in late stage rifting, and melt in the lower crust likely feeds magmatic activity. The crust between the location of the Miocene Red Sea rift axis and the current rift axis is thinner (<22 km) with higher V P /V S (>2.0) than beneath the eastern part of Afar (>26 km, V P /V S < 1.9). This suggests that the eastern region contains less partial melt, has undergone less stretching/extension and has preserved a more continental crustal signature than west of the current rift axis. The Red Sea rift axis appears to have migrated eastward through time to accommodate the migration of the Afar triple junction.

S U M M A R YThe northern Ethiopian rift forms the third arm of the Red Sea, Gulf of Aden triple junction, and marks the transition from continental rifting in the East African rift to incipient oceanic spreading in Afar. We determine the P-and S-wave velocity structure beneath the northern Ethiopian rift using independent tomographic inversion of P-and S-wave relative arrival-time residuals from teleseismic earthquakes recorded by the Ethiopia Afar Geoscientific Lithospheric Experiment (EAGLE) passive experiment using the regularised non-linear least-squares inversion method of VanDecar. Our 79 broad-band instruments covered an area 250 × 350 km centred on the Boset magmatic segment ∼70 km SE of Addis Ababa in the centre of the northern Ethiopian rift. The study area encompasses several rift segments showing increasing degrees of extension and magmatic intrusion moving from south to north into the Afar depression. Analysis of relative arrival-time residuals shows that the rift flanks are asymmetric with arrivals associated with the southeastern Somalian Plate faster (∼0.65 s for the P waves; ∼2 s for the S waves) than the northwestern Nubian Plate. Our tomographic inversions image a 75 km wide tabular low-velocity zone (δV P ≈ −1.5 per cent, δV S ≈ −4 per cent) beneath the less-evolved southern part of the rift in the uppermost 200-250 km of the mantle. At depths of >100 km, north of 8.5 • N, this low-velocity anomaly broadens laterally and appears to be connected to deeper low-velocity structures under the Afar depression. An off-rift low-velocity structure extending perpendicular to the rift axis correlates with the eastern limit of the E-W trending reactivated Precambrian Ambo-Guder fault zone that is delineated by Quaternary eruptive centres. Along axis, the low-velocity upwelling beneath the rift is segmented, with low-velocity material in the uppermost 100 km often offset to the side of the rift with the highest rift flank topography. Our observations from this magmatic rift zone, which is transitional between continental and oceanic rifting, do not support detachment fault models of lithospheric extension but instead point to strain accommodation via magma assisted rifting.

Abstract. Bakaninbreen, southern Svalbard, began a prolonged surge during 1985. In 1986, an internal reflecting horizon on radio echo sounding data was interpreted to show that the position of the surge front coincided with a transition between areas of warm (unfrozen) and cold (frozen) bed. Ground-penetrating radar lines run in 1996 and 1998 during early quiescence show that the basal region of the glacier is characterized by a strong reflection, interpreted as the top of a thick layer of sediment-rich basal ice. Down glacier of the present surge front, features imaged beneath the basal reflection are interpreted as the bottom of the basal ice layer, the base of a permafrost layer, and local ice lenses. This indicates that this region of the bed is cold. Up glacier of the surge front, a scattering zone above the basal reflection is interpreted as warm ice. There is no evidence for this warm zone down glacier of the surge front, nor do we see basal permafrost up glacier of it. Thus, as in early surge phase, the location of the surge front is now at the transition between warm and cold ice at the glacier bed. We suggest that the propagation of the front is associated with this basal thermal transition throughout the surge. Because propagation of the front occurs rapidly and generates only limited heat, basal motion during fast flow must have been restricted to a thin layer at the bed and occurred by sliding or deformation localized at the ice-bed interface.

The northern Main Ethiopian Rift captures the crustal response to the transition from continental rifting in the East African rift to the south, to incipient seafloor spreading in the Afar depression to the north. The region has also undergone plume-related uplift and flood basalt volcanism. Receiver functions from the EAGLE broadband network have been used to determine crustal thickness and average Vp/Vs for the northern Main Ethiopian Rift and its flanking plateaus.On the flanks of the rift, the crust on the Somalian plate to the east is 38 to 40 km thick. On the western plateau, there is thicker crust to the NW (41–43 km) than to the SW (<40 km); the thinning taking place over an off-rift upper mantle low-velocity structure previously imaged by travel-time tomography. The crust is slightly more mafic (Vp/Vs ∼ 1.85) on the western plateau on the Nubian Plate than on the Somalian Plate (Vp/Vs ∼ 1.80). This could either be due to magmatic activity or different pre-rift crustal compositions. The Quaternary Butajira and Bishoftu volcanic chains, on the side of the rift, are characterized by thinned crust and a Vp/Vs > 2.0, indicative of partial melt within the crust.Within the rift, the Vp/Vs ratio increases to greater than 2.0 (Poisson’s ratio, σ > 0.33) northwards towards the Afar depression. Such high values are indicative of partial melt in the crust and corroborate other geophysical evidence for increased magmatic activity as continental rifting evolves to oceanic spreading in Afar. Along the axis of the rift, crustal thickness varies from around 38 km in the south to 30 km in the north, with most of the change in Moho depth occurring just south of the Boset magmatic segment where the rift changes orientation. Segmentation of crustal structure both between the continental and transitional part of the rift and on the western plateau may be controlled by previous structural inheritances. Both the amount of crustal thinning and the mafic composition of the crust as shown by the observed Vp/Vs ratio suggest that the magma-assisted rifting hypothesis is an appropriate model for this transitional rift.