On 31 August a new eruption began from the same fissure and is still ongoing at the time of writing. After 4 September the movement associated with the dyke was minor, suggesting an approximate equilibrium between inflow of magma into the dyke and magma flowing out of it feeding the eruption. Minor eruptions may have occurred under Vatnajškull; shallow ice depressions marked by circular crevasses (ice cauldrons) were discovered in the period 27/08-07/09, indicating leakage of magma or magmatic heat to the glacier causing basal melting ( Fig. 1 and 2b). On 5 September, aircraft radar profiling showed that the ice surface in the centre of the B ‡r!arbunga caldera had subsided 16 m relative to the surroundings, resulting in a 0.32±0.08 km 3 subsidence bowl ( can be compared to a 1 day interferogram over the ice surface spanning 27 -28 August (Fig. 1), that has maximum line-of-sight (LOS) increase of 57 cm, indicating 55-70 cm of subsidence, during 24 hours. From 24 August to 6 September 16 M≥5 earthquakes occurred on the caldera boundary.Over 22000 earthquakes were automatically detected 16/08-06/09 2014, 5000 of which have been manually checked. Four thousand of these have been relatively relocated, defining the dyke segments. Ground deformation in areas outside the Vatnajškull ice cap, and on nunataks within the ice cap, is well mapped by a combination of InSAR, continuously recording GPS sites, and campaign GPS measurements. The GPS observations and analysis give the temporal evolution of the three-dimensional displacements used in the modelling (Fig. 1). Interferometric analysis of synthetic aperture radar images from the COSMO-SkyMed, RADARSAT-2 and TerraSAR-X satellites was used to form 11 interferograms showing LOS change spanning different time intervals (Supplementary Fig. 2). The analysis of seismic and geodetic data is described in Methods.Initial modelling of the dyke, with no a priori constraints on position, strike or dip, show the deformation data require the dyke to be approximately vertical and line up with the seismicity (Extended Data item 4). We therefore fixed the dip to be vertical and the lateral position of the dyke to coincide with the earthquake locations.We modelled the dyke as a series of rectangular patches and estimated the opening and slip on each patch ( Fig. 3a; see Supplementary Figures 3-4 for slip and standard deviations of opening). We used a Markov-chain Monte Carlo approach to estimate 7 the multivariate probability distribution for all model parameters (Methods) on each day 16/08-06/09 2014 (Fig. 2d). The results suggest that most of the magma injected into the dyke is shallower than the seismicity, which mostly spans the depth range from 5 to 8 km below sea level (see Fig. 2c and Methods). While magma may extend to depths greater than 9 km near the centre of the ice cap, towards the edge of the ice cap where constraints from InSAR and GPS are much better, significant opening is all shallower than 5 km (Fig. 3a). The total volume intruded into the dyke by 28 August was 0.48-0...

Gradual inflation of magma chambers often precedes eruptions at highly active volcanoes. During such eruptions, rapid deflation occurs as magma flows out and pressure is reduced. Less is known about the deformation style at moderately active volcanoes, such as Eyjafjallajökull, Iceland, where an explosive summit eruption of trachyandesite beginning on 14 April 2010 caused exceptional disruption to air traffic, closing airspace over much of Europe for days. This eruption was preceded by an effusive flank eruption of basalt from 20 March to 12 April 2010. The 2010 eruptions are the culmination of 18 years of intermittent volcanic unrest. Here we show that deformation associated with the eruptions was unusual because it did not relate to pressure changes within a single magma chamber. Deformation was rapid before the first eruption (>5 mm per day after 4 March), but negligible during it. Lack of distinct co-eruptive deflation indicates that the net volume of magma drained from shallow depth during this eruption was small; rather, magma flowed from considerable depth. Before the eruption, a ∼0.05 km(3) magmatic intrusion grew over a period of three months, in a temporally and spatially complex manner, as revealed by GPS (Global Positioning System) geodetic measurements and interferometric analysis of satellite radar images. The second eruption occurred within the ice-capped caldera of the volcano, with explosivity amplified by magma-ice interaction. Gradual contraction of a source, distinct from the pre-eruptive inflation sources, is evident from geodetic data. Eyjafjallajökull's behaviour can be attributed to its off-rift setting with a 'cold' subsurface structure and limited magma at shallow depth, as may be typical for moderately active volcanoes. Clear signs of volcanic unrest signals over years to weeks may indicate reawakening of such volcanoes, whereas immediate short-term eruption precursors may be subtle and difficult to detect.

[1] GPS observations in south Iceland between 1994 and 2003 are compared with twodimensional elastic half-space and viscoelastic coupling models for two parallel rift zones, representing the Western and Eastern volcanic zones (WVZ, EVZ). GPS data from the Hreppar block, between the WVZ and EVZ, fit a rigid block model within uncertainties. Spreading rates across the WVZ increase from 2.6 ± 0.9 mm/yr in the northeast to 7.0 ± 0.4 mm/yr in the southwest. Conversely, spreading rates in the EVZ decrease from 19.0 ± 2.0 mm/yr in the northeast to 11.0 ± 0.8 mm/yr in the southwest, the direction of ridge propagation. Summed extension rates across the two rift zones are approximately constant and equal to the total plate rate, $18-20 mm/yr, consistent with a simple propagating ridge model whereby the WVZ is deactivating in the direction of EVZ propagation. The coupling model confirms results from the simple elastic half-space model, including relatively shallow locking depths (<5 km) beneath the rift zones, and allows for an estimate of mean viscosity ($10 19

S U M M A R YIceland is one of the few places on Earth where a divergent plate boundary can be observed on land. Direct observations of crustal deformation for the whole country are available for the first time from nationwide Global Positioning System (GPS) campaigns in 1993 and 2004. The plate spreading across the island is imaged by the horizontal velocity field and high uplift rates (≥10 mm yr −1 ) are observed over a large part of central and southeastern Iceland. Several earthquakes, volcanic intrusions and eruptions occurred during the time spanned by the measurements, causing local disturbances of the deformation field. After correcting for the largest earthquakes during the observation period, we calculate the strain rate field and find that the main feature of the field is the extension across the rift zones, subparallel to the direction of plate motion. Kinematic models of the horizontal plate spreading signal indicate a slightly elevated rate of spreading in the Northern Volcanic Zone (NVZ) (23 ± 2 mm yr −1 ), while the rates at the other plate boundary segments agree fairly well with the predicted rate of plate spreading (∼20 mm yr −1 ) across Iceland. The horizontal ISNET velocities across north Iceland therefore indicate that the excessive spreading rate (>30 mm yr −1 ) observed by GPS in 1987-1992 following the 1975-1984 Krafla rifting episode was significantly slower during 1993-2004. We model the vertical velocities using glacial isostatic adjustment (GIA) due to the recent thinning of the largest glaciers in Iceland. A layered earth model with a 10-km thick elastic layer, underlain by a 30-km thick viscoelastic layer with viscosity 1 × 10 20 Pa s, over a half-space with viscosity ∼1 × 10 19 Pa s can explain the broad area of uplift in central and southeastern Iceland. A wide area of significant residual uplift (up to 8 mm yr −1 ) is evident in north Iceland after we subtract the rebound signal from the observed rates, whereas the Reykjanes Peninsula and the Western Volcanic Zone (WVZ) appear to be subsiding at a rate of 4-8 mm yr −1 . We observe a coherent pattern of small but significant residual horizontal motion (up to 3 mm yr −1 ) away from Vatnajökull and the smaller glaciers that is most likely caused by glacial rebound. Our study demonstrates that the velocity field over a large part of Iceland is affected by deglaciation and that this effect needs to be considered when interpreting deformation data to monitor subglacial volcanoes in Iceland.

No abstract

We invert geodetic measurements of coseismic deformation from the 1989 MS7.1 Loma Prieta earthquake to determine the geometry of the fault and the distribution of slip on the fault plane. The data include electronic distance measurements, Global Positioning System and very long baseline interferometry vectors, and elevation changes derived from spirit leveling. The fault is modeled as a rectangular dislocation surface in a homogeneous, elastic half‐space. First, we assume that the slip on the fault is uniform and estimate the position, orientation, and size of the fault plane using a nonlinear, quasi‐Newton algorithm. The best fitting dislocation strikes N48°±4°W and dips 76°±9°SW, consistent with the trend of the aftershock zone and moment tensor solutions. Bootstrap resampling of the data is used to graphically illustrate the uncertainty in the location of the rupture plane. The 95% confidence envelope overlaps the aftershock zone, arguing that there is not a significant discrepancy between the geodetic data and the aftershock locations. Second, we estimate the slip distribution using the best fitting uniform slip fault orientation but increase the fault length to 40 km and the downdip width to 18 km. The fault is divided into 162 subfaults, 18 along strike and 9 along dip. Each subfault is allowed to have constant right‐lateral and reverse components of slip. We then solve for the slip on each subfault that minimizes a linear combination of the norm of the weighted data residual and the roughness of the slip distribution. The smoothing parameter, which determines the relative weight put on fitting the data versus smoothing the slip distribution, is chosen by cross validation. Simulations indicate that cross‐validation estimates of the smoothing parameter are nearly optimal. The preferred slip distribution is very heterogeneous, with maximum strike slip and dip slip of about 5 and 8 m, respectively, located roughly 10 km north of the hypocenter. There is insignificant dip slip in the southeastern most part of the fault, causing the rake to vary from nearly pure right‐lateral in the southeast to oblique right‐reverse in the northwest. The change in rake is consistent with a uniform stress field if the fault dip increases by about 10° toward the southeast, as indicated by the aftershock locations. There was little slip above 4 km depth, consistent with the observation that there was little, if any, surface rupture.

We analyze data spanning up to 5 years from 18 continuous GPS stations in Iceland, computing daily positions of the stations with three different high‐level geodetic processing software packages. We observe large‐scale crustal deformation due to plate spreading across Iceland. The observed plate divergence between the North American and the Eurasian plates is in general agreement with existing models of plate motion. Spreading is taken up within a ∼100–150 km wide plate boundary zone that runs through the island. Of the two parallel branches of the plate boundary in south Iceland, the eastern volcanic zone is currently taking up the majority of the spreading and little is left for the western volcanic zone. The plate boundary deformation field has been locally and temporarily affected in south Iceland by two Mw = 6.5 earthquakes in June 2000, inflation at Katla volcano during 2000 to 2004, and an eruption of Hekla volcano in February 2000. All stations with significant vertical velocities are moving up relative to the reference station REYK, with the highest velocity exceeding 20 mm/yr in the center of the island.

Magma flow during volcanic eruptions causes surface deformation that can be used to constrain the location, geometry and internal pressure evolution of the underlying magmatic source 1 . The height of the volcanic plumes during explosive eruptions also varies with magma flow rate, in a nonlinear way 2,3 . In May 2011, an explosive eruption at Grímsvötn Volcano, Iceland, erupted about 0.27 km 3 denserock equivalent of basaltic magma in an eruption plume that was about 20 km high. Here we use Global Positioning System (GPS) and tilt data, measured before and during the eruption at Grímsvötn Volcano, to show that the rate of pressure change in an underlying magma chamber correlates with the height of the volcanic plume over the course of the eruption. We interpret ground deformation of the volcano, measured by geodesy, to result from a pressure drop within a magma chamber at about 1.7 km depth. We estimate the rate of magma discharge and the associated evolution of the plume height by differentiating the co-eruptive pressure drop with time. The time from the initiation of the pressure drop to the onset of the eruption was about 60 min, with about 25% of the total pressure change preceding the eruption. Near-real-time geodetic observations can thus be useful for both timely eruption warnings and for constraining the evolution of volcanic plumes.