Volcanoes in Iceland and Crustal Deformation Processes
- Central Volcano
Focal point of eruptive activity within a system. Often constructed by repeated eruptions from a central vent system that is maintained by a long-lived magmatic plumbing system.
Magma body with thickness much smaller than its lateral extent. Formed by magma intruding into cracks.
- Dry Tilt
Dry-tilt measurements are high-precision optical leveling measurements performed on short lines oriented EW and NS or in a circle.
- Fissure Swarm
High-density subparallel volcanic fissures and tectonic fractures extending from a specific volcano.
The study of the shape and area of the Earth, from large-scale variations that affect the rotation dynamics of the planet down to smaller length scales of earthquakes, landslides, etc.
Glacial isostatic adjustment. The adjustment of the crust after an ice load is removed from the surface. Here GIA is defined as the signal composed of the short-term elastic response and long-term viscoelastic adjustment due to the retreat of the ice caps in Iceland.
Global Navigation Satellite System. Satellite-based positioning systems, such as GPS, GLONASS, and GALILEO, with a network of satellites that transmit signals which can be used to infer three-dimensional positions.
US Global Positioning System. A network of satellites that transmit signals which can be used to infer three-dimensional positions. The first GPS satellite was launched in 1978. The system had 24 satellites (full constellation) available by the end of 1993 and was declared fully operational in 1995.
Interferometric Synthetic Aperture Radar. The combination of Synthetic Aperture Radar imagery (generally acquired from airborne or satellite-based platforms) to infer ground deformation, digital elevation models, variations in atmospheric water vapor, etc.
Sudden glacial outburst flood.
Measurements of geodetic height relative to a reference surface. Optical leveling instrument are used to measure height between benchmarks in a line.
- Magma Chamber
Reservoir of molten rock (magma) in the crust.
Linear zone where the Earth’s crust is being pulled apart due to extensional tectonics.
Tabular sheet intrusion, where magma has intruded between horizontal layers of rock.
- Volcanic System
Characterized by a central volcano, and often fissure swarm, a caldera, and high-temperature geothermal area associated with an upper crustal magma chamber.
Definition of the Subject
Crustal deformation processes associated with active volcanoes can vary from one volcano to another and from one period to the next. Measurements of crustal deformation can provide important information on processes at depth and the state of a volcano. For a single reservoir system, pressure increases as magma accumulates in the reservoir, with gradual buildup of stress in the roof of the magma chamber. When a critical limit is reached, the wall of the magma chamber fails followed by intrusion or eruption. The preeruptive inflation and syn-eruptive deflation can be used to evaluate source depth and magma volume. The depth and the shape of the magma reservoir vary from one volcano to the next, and, at some volcanoes, the geodetic data suggest more than one chamber. The rate at which magma migrates into the reservoir can be episodic, and the effects of tectonic stresses and loads can change with time and evolve during volcanic events.
Many volcanoes are continuously deforming in response to internal process such as magma movements. Others are fairly quiet. Repeated geodetic measurements along with seismic monitoring are the key to better understanding volcanic processes and mitigating risk. Geodetic results from Icelandic volcanoes range from zero deformation or subsidence to continuously recharging magmatic systems (Dvorak and Dzurisin 1997; Massonnet and Sigmundsson 2000; Sigmundsson 2006; Segall 2010).
Iceland is located on the Mid-Atlantic Ridge where the rate of spreading between the North American and the Eurasian plate is about 18–19 mm/yr (DeMets et al. 2010). There are about 30 active volcanic systems in Iceland, and historically there has been on average an eruption every 4–5 years (Thordarson and Larsen 2007). The anomalously high volcanic activity and crustal thickness in Iceland has been explained by the existence of a mantle plume (Wolfe et al. 1997; Allen et al. 2002) interacting with the ridge. Iceland’s existence above sea level makes it an ideal place to study crustal deformation and volcanic activity at a divergent plate boundary.
The Reykjanes Ridge (RR) emerges onshore at the southwest tip of the oblique spreading Reykjanes Peninsula (RP) (Hreinsdóttir et al. 2001; Keiding et al. 2008). At the Hengill triple junction, the RP connects to the West Volcanic Zone (WVZ) and the South Iceland Seismic Zone (SISZ). The SISZ transforms the main rift axis to the East Volcanic Zone (EVZ) (Sigmundsson et al. 1995), which is currently the volcanically most active region in Iceland (Thordarson and Larsen 2007). The full plate spreading in South Iceland is currently taken up over both the EVZ and the less active WVZ (La Femina et al. 2005; Geirsson et al. 2012). In north Iceland the active plate spreading takes place across the North Volcanic Zone (NVZ) (Pedersen et al. 2009; Drouin et al. 2014). The Tjörnes Fracture Zone (TFZ) shifts the rift system offshore to the Kolbeinsey Ridge (Rögnvaldsson et al. 1998; Metzger et al. 2011).
Three volcanic belts are located outside the main rift zone (Jakobsson 1972; Sæmundsson 1978; Óskarsson et al. 1985; Einarsson 2008). The Snæfellsnes Volcanic Belt in West Iceland is an old rift zone where volcanism was reactivated 2 Ma ago (Jóhannesson 1982). The South Iceland Volcanic Belt (SIB) is the southwestward propagating continuation of the EVZ but with little or no active rifting (Geirsson et al. 2012) and extends offshore to Vestmannaeyjar islands. The Öræfajökull Volcanic Belt is mostly subglacial and east of the current rift axis. Three volcanic systems have been identified in the volcanic belt, but only Öræfajökull volcano, the highest mountain in Iceland (2110 m), has verified eruptions during Holocene (Thordarson and Larsen 2007).
About 70% of all historical eruptions originate from the Grímsvötn, Hekla, Katla, and the Bárdarbunga-Veidivötn volcanic systems. The most active volcanic system in Iceland is Grímsvötn, with over 60 documented eruptions in historical times (Larsen et al. 1998). The Grímsvötn fissure system extends about 90 km SW from the subglacial central volcano (Fig. 2). In 1783–1784 the Laki fires took place along a 27 km-long ice-free segment of the fissure swarm, producing about 14.7 km3 of lava (Thordarson and Self 1993). The Katla volcanic system in South Iceland has 21 confirmed historic eruptions, including the Eldgjá fissure eruption in 934 AD that produced the largest volume of lava erupted in historical times, about 20 km3 (Thordarson et al. 2001; Sigurdardóttir et al. 2015). The central volcano has a 110 km2 and 700 m deep, ice-covered caldera. The other 20 historical Katla eruptions were phreatomagmatic basalt eruptions within the caldera, generating jökulhlaups. Hekla volcanic system has 23 documented historical eruptions, the majority of which have been centered on the Hekla central volcano. The Hekla eruptions are mixed, where silica content of the first erupted material is dependent on the repose time since the previous eruption (Thordarson and Larsen 2007). The 190 km-long Bárdarbunga-Veidivötn system has erupted at least 27 times in the historical period, including both phreatomagmatic and fissure eruptions (Larsen and Gudmundsson 2014). On 31 August 2014, a fissure eruption started NE of the Bárdarbunga central volcano, after a 2-week seismic unrest and dike propagation over 45 km distance from the subglacial volcanic center (Sigmundsson et al. 2015). The eruption lasted until 27 February 2015 with 1.6 ± 0.3 km3 of lava erupted (Gíslason et al. 2015), making it the largest eruption in Iceland since the 1783–1784 Laki eruption. During the eruption the ice-filled Bárdarbunga caldera subsided 65 m due to withdrawal of magma from a reservoir (Gudmundsson et al. 2016).
Volcano Geodesy in Iceland
Repeated geodetic measurements have been used to monitor and study volcano deformation processes in Iceland since the 1960s. In 1938, a triangulation network was set up across the NVZ centered at Krafla volcano, to test Wegener’s continental drift hypothesis (Niemczyk 1943; Tryggvason 1984). The 100 km-long network was not measured again until 1965, using electronic distance measurements (EDM) and repeated every few years for the next two decades alongside leveling profiles (Gerke 1974; Wendt et al. 1985). Leveling lines and optical leveling tilt stations (dry tilt) were established at several volcanoes in the 1960s to 1980s and measured regularly for monitoring purposes (e.g., Tryggvason 1968, 1989, 1994a, b, 1995; Sturkell et al. 2006). Watertube tilt meters were used at Hekla from 1968 to 1970 but were discontinued when it became apparent that dry-tilt measurements were of similar precision and easier to perform (Tryggvason 1994b). A 69 m-long watertube tilt meter was set up at the Krafla Power Plant in August 1976 to monitor deformation during the Krafla fires between 1975 and 1984. In 1977, continuously recording electronic tilt meters were installed in the Krafla region, which obtained higher precision and more frequent measurements (Tryggvason 1982). However, electronic meters are sensitive to temperature and drift slowly with time (Tryggvason 1995). In addition to more specialized geodetic equipment, a tape measure and creep meters were used during rifting episodes in Krafla to measure changes in fissure opening (e.g., Hauksson 1983). Lakes have been used as natural tilt meters, e.g., lake Öskjuvatn in Askja and lake Mývatn, SSE of Krafla volcano (Tryggvason 1987, 1989; Björnsson and Eysteinsson 1998). In late 1979, a network of borehole strain meters to monitor Hekla was installed 15 to 45 km from the volcano (Linde et al. 1993).
During the past 25 years, Global Navigation Satellite System (GNSS) and Interferometric Synthetic Aperture Radar (InSAR) space geodetic techniques have been widely used to monitor and study volcanic deformation. GNSS geodetic measurements were first conducted in Iceland in 1986 when a 51 site, countrywide campaign network was established and measured using GPS data (US Global Positioning System) (Foulger et al. 1987). The measurements determine the location of a point on the ground with sub-centimeter level accuracy, allowing us to monitor the deformation of the surface with time. The first GNSS geodetic network to monitor volcanic activity was measured around Hekla volcano in 1989 (Sigmundsson et al. 1992). Several hundred geodetic benchmarks are now measured annually in Iceland. The first two continuous GNSS stations in Iceland were the International GNSS Service (IGS) stations Reykjavík (REYK), installed in 1995, and Höfn (HOFN), in 1997. In response to volcanic unrest, six GNSS stations were installed in 1999 and 2000 to monitor the Hengill, Eyjafjallajökull, and Katla volcanoes (Geirsson et al. 2006). The continuous GNSS network has grown rapidly in the last 10 years. In January 2014 there were 90 GNSS stations operating in Iceland, monitoring and measuring crustal deformation, the majority of which are located within the active neo-volcanic zone (Fig. 2). During periods of volcanic unrest, the GNSS network is augmented with semicontinuous stations (Hreinsdóttir et al. 2012; Sigmundsson et al. 2015).
InSAR measurements have been used to study volcanic deformation since 1992, after the launch of the first European Space Agencies (ESA) satellite, ERS1 (Massonnet et al. 1995; Massonnet and Sigmundsson 2000). By taking the phase difference between two SAR images, an interferogram can be constructed to reveal the surface deformation occurring in the time between two image acquisitions. When coherence is good, InSAR measurements provide a spatial coverage and resolution that can play an important role in understanding the nature of surface deformation (e.g., Ófeigsson et al. 2011a).
InSAR measurements to study volcanic processes in Iceland are mostly limited to the relatively ice-free volcanoes and conducted during the summer months. However, during the 2014 Bárdarbunga eruption, InSAR data were used along with airplane radar profiling and GNSS measurements to monitor the rapid subsidence of the ice-covered Bárdarbunga caldera (Sigmundsson et al. 2015; Gudmundsson et al. 2016). InSAR techniques have also been used to study surface processes such as ice melting due to volcano-glacier interaction (Björnsson et al. 2001).
Deformation in Iceland
Crustal deformation in Iceland is caused by multiple sources. In order to study and monitor volcanic deformation, it is important to understand the background signals.
Glaciers and ice caps cover about 11% of the country (Björnsson and Pálsson 2008). The glaciers are melting and retreating, causing the crust to rebound. The Earth’s elastic and viscoelastic response to ice retreat results in uplift rates currently exceeding 35 mm/yr in central Iceland (Compton et al. 2015; Fridriksdóttir 2014; Fig. 3) and horizontal movement away from the ice caps (Árnadóttir et al. 2009; Geirsson et al. 2010; Auriac et al. 2013). Annual load changes due to snow and ice (Grapenthin et al. 2006; Drouin et al. 2016; Compton et al. 2017) cause an elastic response of the crust. Other load changes such as the draining of lakes, surging glaciers (Auriac et al. 2014), or the embedment of new lava flows (e.g., Grapenthin et al. 2010) can result in a significant deformation signal. In addition, young lava flows cool and contract, which can be observed in InSAR images (e.g., Sigmundsson et al. 1997a; Ófeigsson et al. 2011a; Wittmann et al. 2017).
Geothermal utilization of high-temperature geothermal field associated with a volcanic system can result in localized subsidence signal, resulting from pressure change and/or temperature decrease in the geothermal reservoir. Several geothermal power plants are operating in Iceland both extracting geothermal fluids and reinjecting wastewater. In SW Iceland four geothermal power plants are currently in operation, two on Reykjanes Peninsula (Reykjanes and Svartsengi) and two in Hengill (Hellisheiði and Nesjavellir). In North Iceland three geothermal power plants are in operation, at Krafla, Theistareykir, and Bjarnarflag. Deformation related to these power plant has been monitored using leveling, GNSS, and InSAR measurements (e.g., Keiding et al. 2010; Michalczewska et al. 2014; Juncu et al. 2017; Drouin et al. 2017; Parks et al. 2018; Juncu et al. 2018).
Volcano Deformation in Iceland
Over 20 confirmed eruptions have occurred in Iceland since 1970, including 9 during the Krafla fires. Measured crustal deformation leading up to an eruption has varied from a steady inflation of a crustal magma chamber over several years (e.g., Ófeigsson et al. 2011a) to a more complex episodic movement of magma into dikes and sills for a few weeks to years prior to an eruption (Sigmundsson et al. 2010, 2015; Hjaltadóttir et al. 2015). Co-eruptive deformation often involves more than one source, with the formation of feeder dikes and the depletion of magma source(s) at depth (e.g., Sturkell et al. 2013; Hreinsdóttir et al. 2014).
Hekla, a ridge-shaped stratovolcano in South Iceland, has 18 confirmed eruptions since 1104 AD, with 1 to 3 eruptions per century (Thordarson and Larsen 2007). In the past 50 years, the volcano has erupted at a regular and more frequent rate, with relatively small eruptions in 1970, 1980–1981, 1991, and 2000. Hekla eruptions usually start with a short Plinian or subplinian phase followed by lava fountaining and effusive activity (Thordarson and Larsen 2007). The longer the repose time, the larger and more silica rich are the eruptions (Thorarinsson 1967). This suggests a reservoir where magma can accumulate and evolve with time.
Geodetic measurements around Hekla volcano during the last three eruptive cycles have shown a steady rate of inflation between eruptions (Kjartansson and Gronvold 1983; Tryggvason 1994b; Ófeigsson et al. 2011a; Geirsson et al. 2012) and rapid co-eruptive deflation (Tryggvason 1994b; Sigmundsson et al. 1992; Ófeigsson et al. 2011a; Sturkell et al. 2013). The data have been used to constrain the depth of the magma chamber beneath Hekla resulting in a wide range of estimates, from 5–6 km (Tryggvason 1994b) to 22–29 km depth (Geirsson et al. 2012). Co-eruptive strain measurements indicate a reservoir at around 10 km depth (Sturkell et al. 2013), but recent GPS and InSAR data suggest a more deeply seated magma chamber (over 14 km depth) (Geirsson et al. 2012; Ófeigsson et al. 2011a), in agreement with seismic observations (Soosalu and Einarsson 2004). Both co-eruptive InSAR and strain meter data indicate a shallow feeding dike that does not extend all the way to the reservoir (Sturkell et al. 2013). During early surveys of the volcano, far-field and near-field tilt measurements sometimes gave contradicting results (e.g., Tryggvason 1994b). This discrepancy could possibly have originated from the shallow dike feeding the eruption. In addition, a localized subsidence signal is observed in InSAR images around recent lava flows around the summit of Hekla (Ófeigsson et al. 2011a). The subsidence is caused by both thermal contraction of the lava itself and loading effects (Ófeigsson et al. 2011a, Grapenthin et al. 2010; Wittmann et al. 2017). The extent of the signal is limited, but it can bias estimates of source parameters.
The Katla volcano lies under the Mýrdalsjökull ice cap in South Iceland. It has 20 confirmed phreatomagmatic basalt eruptions within the ice-filled caldera over the past 1100 years (Thordarson and Larsen 2007). A low-velocity anomaly and a clear S-wave shadow beneath the Katla caldera have been interpreted in terms of a magma chamber at shallow depth (Gudmundsson et al. 1994). Katla is seismically the most active volcano in Iceland, with periods of elevated unrest since 1952, most recently in 1999–2004 (Sturkell et al. 2008b) and 2011–2012 (Ófeigsson et al. 2011b). The seismic activity also shows annual variability, with activity peaking in the fall, possibly related to elevated fluid pressure and reduced ice load (Einarsson and Brandsdóttir 2000). Snow thickness variations of up to 6 m have also been considered a triggering factor for seasonal seismicity as well as eruptions (Albino et al. 2010). Interestingly, documented Katla eruptions tend to peak in the late summer and fall, with no documented eruption starting from December through April (Thordarson and Larsen 2007).
Since the 1100s, one to three Katla eruptions have occurred per century, each accompanied by a jökulhlaup. In contrast to observations at Hekla volcano, the magnitudes of Katla eruptions do not appear to correlate with the length of the preceding repose interval. However, longer eruption intervals tend to follow large eruptions (Elíasson et al. 2006). The most recent Katla eruption took place in 1918. It was a big eruption, accompanied by a massive glacial outburst flood (e.g., Thordarson and Larsen 2007 and reference therein). Three smaller events, increased seismic activity accompanied by a small glacial outburst floods, occurred in 1955, 1999, and 2011. It has been suggested that these floods were caused by small subglacial eruptions or shallow magma intrusions under the glacier (Einarsson and Brandsdóttir 2000; Ófeigsson et al. 2011b).
Geodetic studies of Katla volcano are limited to the volcano flank, outside the Mýrdalsjökull ice cap, and nunataks at the caldera rim. In 1966, three dry-tilt stations were established east of Mýrdalsjökull glacier to monitor the volcano. Measurements from 1967 to 1971 showed repeated uplift and subsidence that correlated with snow load variations on the glacier (Tryggvason 1973). GPS campaign measurements outside the glacier have shown steady uplift and deformation away from the center of the volcano since measurements started in 1992. Two GNSS stations have been operating at the southern flank of the volcano since 1999, and another was set up over an existing benchmark on the western flank of the volcano in 2006. The sites show a combination of annual variations due to load changes (Grapenthin et al. 2006; Pinel et al. 2007) and a steady uplift signal. To evaluate how much of the observed uplift was due to glacial retreat in the region rather than of volcanic origin, measurements of glacial ice thickness variations were used to predict the response of the crust to the glacier thinning (Pinel et al. 2007; Sturkell et al. 2008b). The models suggested that the observed horizontal measurements around Katla were too large compared to the vertical data to be caused by rebound only, thus concluding that the data were consistent with magma accumulation beneath the volcano. The study, however, only considered the contribution of the Mýrdalsjökull ice cap in the calculations. InSAR time series spanning 2003 to 2009 from South Iceland show no sign of inflation around Katla volcano after being corrected for GIA model calculations (Spaans 2011; Spaans et al. 2015).
Campaign GPS data from nunataks within the Mýrdalsjökull glacier suggested rapid inflation during the 2000 to 2004 period (Sturkell et al. 2008b), which along with seismic unrest was considered a sign that Katla was preparing for an eruption. In 2010, continuous GNSS stations were set up at the nunataks, revealing a complex deformation signal presumably originating from within the caldera (Ófeigsson et al. 2011b; Spaans et al. 2015). The signal is annual with up to 60 mm variations in the horizontal and 70 mm in the vertical. Any measurement at the nunataks would therefore have been highly dependent on their timing.
Leveling and tilt measurements at Krafla volcano provided important information both for monitoring and forecasting rifting events during the Krafla fires and to study the dynamics of rifting processes (e.g., Buck et al. 2006; Wright et al. 2012). Continuous uplift with gradually increasing seismic activity within the caldera for a few weeks to months was followed by a rapid subsidence and volcanic tremor from a couple of hours to several days, often associated with an earthquake swarm in the fissure swarm and increased geothermal activity (e.g., Björnsson et al. 1979; Einarsson 1991). The initial rate of uplift of the Krafla caldera following each rifting event was about 6 mm/day but then slowed down with earthquake activity increasing steadily. The observations are consistent with a magma flow rate of about 5 m3/s into a 3 km deep magma chamber (Björnsson et al. 1979) from a deeper reservoir (Tryggvason 1986; Árnadóttir et al. 1998). When a critical pressure was reached, the magma extruded laterally out of the magma chamber and into the fissure swarm, where widening of fissures and subsidence was observed. A maximum accumulated widening of about 8–9 m was measured from 1975 to 1984 across the central part of the fissure system (Tryggvason 1984, 1994a), corresponding to 500 years of long-term plate spreading.
During 1980–1981, four eruptions took place along fissures extending 10 km north of the caldera. The last rifting event took place in September 1984 with the largest eruption of the sequence, an 8.5 km-long fissure eruption with 0.11 km3 of lava (dense rock equivalent) erupted (Rossi 1997). During the Krafla fires, only a small portion of the estimated magma leaving the shallow magma chamber erupted at the surface. After the eruption in 1984 ended, inflation continued for about 5 years (Tryggvason et al. 1994a). Since 1989 a steady subsidence has been observed in the region (Fig. 8).
InSAR and GPS measurements, along with leveling and tilt measurements, have been conducted regularly around Krafla since the end of the Krafla fires (e.g., Foulger et al. 1992; Sigmundsson et al. 1997a; Sturkell et al. 2008a; de Zeeuw-van Dalfsen et al. 2004). The data show subsidence along the fissure swarm and continued deflation of the magma chamber, with rate decreasing with time. In addition localized subsidence signal around boreholes has been observed at the Krafla power station (Drouin et al. 2017). A broad signal of uplift seen in InSAR images spanning 1993–1999 has been interpreted as a deep-seated magma accumulation at the volcano (de Zeeuw-van Dalfsen et al. 2004).
Bárdarbunga volcano in central Iceland lies beneath the northwestern part of the Vatnajökull ice cap. It rises over 2000 meters above sea level and has an 80 km2 ice-filled caldera. The Bárdarbunga-Veidivötn fissure swarm extends 115 km SSW and 55 km NNE from the central volcano (Thordarson and Larsen 2007).
On 29 September 1996 at 10:48 UTC, an Mw 5.6 event occurred at the northern rim of Bárdarbunga caldera, followed by an intense earthquake swarm (Einarsson et al. 1997; Pagli et al. 2007). The earthquakes migrated 20 km south toward the Grímsvötn volcano over the next 24 h. In the evening of 30 September, a subglacial eruption, the Gjálp eruption, began between the two volcanic centers (Einarsson et al. 1997; Gudmundsson et al. 2004). The eruption took place on a 6 km-long, NS trending fissure, initially below 550 to 750 m thick ice (Gudmundsson et al. 2004). On 2 October the eruption melted an opening through the ice resulting in a phreatomagmatic eruption that lasted until 13 October.
On 16 August 2014, an intense seismic swarm began at Bárdarbunga volcano (Sigmundsson et al. 2015). The earthquakes migrated from the Bárdarbunga caldera in segments with variable strike over 45 km distance for nearly 2 weeks. GPS observations show both deflation of the volcano and displacement consistent with the growth of the dike (Fig. 9). On 24 August a series of M5+ earthquakes started around the Bárdarbunga caldera rim. Airborne radar profiling data and 1-day SAR interferograms over the ice surface revealed up to about 1 m/day subsidence of the caldera floor.
The propagation of the dike was monitored with seismic data, continuous GPS measurements, and InSAR data (Sigmundsson et al. 2015). Significant extension was observed over 100 km distance ESE and WNW of the volcano (Ófeigsson et al. 2015). On 29 August a minor fissure erupted for a few hours in Holuhraun, northeast of Bárdarbunga. On 31 August the eruption started again with a 1.5 km-long fissure erupting. Widening across the dike continued at decaying rate for the first week of September, but GNSS stations continued to respond to the deflating Bárdarbunga volcano (Sigmundsson et al. 2015; Gudmundsson et al. 2016). The eruption ended on the 27 February 2015 with 1.6 ± 0.3 km3 of lava erupted (Gíslason et al. 2015). During the Bárdarbunga unrest and in the months following the eruption, several new GNSS stations were installed to monitor the volcano (Ófeigsson et al. 2015). Following the eruption InSAR and GPS data reveal post-rifting relaxation and isostatic adjustment to the new lava flow (Grapenthin et al. 2018).
The Gjálp 1996 and Holuhraun 2014 eruptions were both initiated by seismic activity at the Bárdarbunga caldera followed by a complex dike propagation and rifting (Einarsson et al. 1997; Pagli et al. 2007; Sigmundsson et al. 2015; Gudmundsson et al. 2016). The events involved neighboring volcanic systems, most significantly the 1996 Gjálp magma having the geochemical signature of Grímsvötn volcanic system. During both 1996 and 2014, triggered seismicity was observed around Tungnafellsjökull volcano (Fig. 9), and InSAR data from 1996 reveal triggered slip on faults up to 30 km distance from Bárdarbunga (Pagli et al. 2007; Parks et al. 2017).
Eyjafjallajökull in South Iceland is an ice-capped volcano with a summit crater, located outside the main rift zone. Only three historical eruptions are known at Eyjafjallajökull prior to 2010, a fissure eruption in 920 and summit eruptions in 1612 and 1821–1823 (Larsen et al. 1999; Óskarsson 2009). On 20 March 2010, a fissure eruption started on the east flank of the volcano followed by an explosive summit eruption on 14 April.
The deformation pattern leading up to the flank eruption was both spatially and temporally variable. During the flank eruption, the deformation almost ceased, and the volcano remained at an inflated state until the eruption ended on 12 April. On 14 April 2010, a more explosive eruption began at the ice-capped summit of the volcano. During this eruption rapid deformation toward the summit and subsidence was observed both with InSAR and at GNSS stations around the volcano (Sigmundsson et al. 2010). The geodetic data are consistent with the deflation of a sill at 4–5 km depth under the summit. The summit eruption ended on 22 May. At around the same time, the far-field sites stopped moving toward the volcano (Hreinsdóttir et al. 2012; Thrastarson 2013).
Following the Eyjafjallajökull eruptions, GNSS stations on the flank of the volcano continued to move into the summit region at a fast decaying rate for a few weeks and months. However, from July 2010 until the end of 2015, the sites closest to the summit moved up and away from the volcano (Fig. 11, Jónsson 2013) suggesting recharging of the shallow magma pocket that erupted during the summit eruption.
Uplift and Subsidence
Several intrusion events in Iceland have been detected with geodetic measurements during the past few decades. From 1994 to 1999, inflation of up to 19 mm/yr in the Hengill volcanic system was observed with GPS, leveling, and InSAR measurements. The data were interpreted in terms of a slow inflow of magma into a reservoir at 7 km depth, resulting in intense seismic activity (Sigmundsson et al. 1997b; Feigl et al. 2000). From 2007 to 2008, deformation and seismic activity was detected in the Kverkfjöll volcanic system, at Mt. Upptyppingar. A dike was intruding into the lower crust, north of Vatnajökull ice cap (Geirsson et al. 2010; Hooper et al. 2011). Inflation was also observed at Theistareykir volcano from 2007 to 2008, with maximum uplift of 30 mm/yr, suggesting magma accumulation in a reservoir at 9 km depth (Metzger et al. 2011). In 2009 inflation began in the Krýsuvík volcanic system, suggesting a pressure source at 4–5 km depth. The region inflated for 9 months but then deflated again, reaching pre-inflation state in April of 2010. Another inflation period started in May, with up to 50 mm/yr uplift rate (Michalczewska et al. 2012). In early 2012 the region started deflating again. Both inflation pulses were associated with increased seismic activity in the crust above the source. Since 2016 Öræfajökull volcano has shown signs of unrest, with increased seismicity and changes in geothermal activity (Jónsdóttir et al. 2018). Subtle signs of inflation have been measured around the volcano, interpreted along with the observed seismic and geothermal activity, as intrusive activity into the volcano (Geirsson et al. 2018).
Geodetic measurements at Krafla volcano show decaying subsidence at the volcano since the Krafla fires, interpreted as cooling and contracting of the magma chamber (Sigmundsson et al. 1997a). Measurements at Askja volcano in north Iceland show rapid deflation since at least 1983 at a slowly decaying rate above a shallow magma chamber (Sturkell and Sigmundsson 2000; Sturkell et al. 2006; de Zeeuw-van Dalfsen et al. 2013). The caldera center subsided almost 1 m from 1983 to 2004. This rapid rate has been interpreted as a pressure drop in the magma chamber, but neither eruption nor known intrusion event took place during this time. GPS measurements show movement into the center of the caldera, and InSAR measurements reveal a few mm/yr subsidence along the fissure swarm (Camitz et al. 1995; Pagli et al. 2006; Pedersen et al. 2009). Askja volcano had a large explosive eruption in 1875, forming lake Öskjuvatn, a 4.5 km-wide caldera. The eruption was a part of a rifting episode in the volcanic system where an approximately 100 km-long segment of the Askja fissure swarm was activated (Sigurdsson and Sparks 1978; Brandsdóttir 1992). Two smaller episodes took place from 1921–1929 to 1961–1962. Subsidence is also observed at Torfajökull (Scheiber 2008; Geirsson et al. 2012), the largest rhyolitic center in Iceland.
- Allen RM, Nolet G, Morgan WJ, Vogfjörd K, Bergsson BH, Erlendsson P, Foulger GR, Jakobsdóttir S, Julian BR, Pritchard M, Ragnarsson S, Stefánsson R (2002) Imaging the mantle beneath Iceland using integrated seismological techniques. J Geohys Res 107:2325. https://doi.org/10.1029/2001JB000595 ADSCrossRefGoogle Scholar
- Björnsson A, Eysteinsson H (1998) Height changes at Krafla 1974–1995 (in Icelandic with English summary), Report OS-98002/NE9801, ISBN 9979-007-5, Orkustofnun, ReykjavíkGoogle Scholar
- Björnsson H, Pálsson F (2008) Icelandic glaciers. Jökull 58:365–386Google Scholar
- Brandsdóttir B (1992) Historical accounts of earthquakes associated with eruptive activity in the Askja volcanic system. Jökull 42:1–12Google Scholar
- Drouin V, Sigmundsson F, Hreinsdóttir S, Ófeigsson BG, Sturkell S, Islam T (2014) Crustal movements at a divergent plate boundary: interplay between volcano deformation, geothermal processes, and plate spreading in the northern volcanic zone, Iceland since 2008. Geophys Res Abstr 16:EGU2014-10740-3Google Scholar
- Einarsson P (2008) Plate boundaries, rifts and transforms in Iceland. Jökull 58:35–58Google Scholar
- Einarsson P, Brandsdóttir B (1980) Seismological evidence for lateral magma intrusion during the July 1978 deflation of the Krafla volcano in NE-Iceland. J Geophys 47:160–165Google Scholar
- Einarsson P, Brandsdóttir B (2000) Earthquakes in the Mýrdalsjökull area, Iceland, 1978–1985: seasonal correlation and connection with volcanoes. Jökull 49:59–74Google Scholar
- Fridriksdóttir HJ (2014) Uplift in the Vatnajökull region estimated with GPS geodetic measurements (in Icelandic), B.Sc. Thesis, University of Iceland, Reykjavík, p 26Google Scholar
- Geirsson H, Árnadóttir T, Hreinsdóttir S, Decriem J, LaFemina PC, Jonsson S, Bennett RA, Metzger S, Holland A, Sturkell E, Villemin T, Volksen C, Sigmundsson F, Einarsson P, Roberts MJ, Sveinbjornson H (2010) Overview of results from continuous GPS observations in Iceland from 1995 to 2010. Jökull 60:3–22Google Scholar
- Geirsson H, LaFemina P, Árnadóttir T, Sturkell E, Sigmundsson F, Travis M, Schmidt P, Lund B, Hreinsdóttir S, Bennett R (2012) Volcano deformation at active plate boundaries: deep magma accumulation at Hekla volcano and plate boundary deformation in South Iceland. J Geophys Res 117:B11409ADSCrossRefGoogle Scholar
- Geirsson H, Parks MM, Ófeigsson BG, Drouin V, Li S, Sigmundsson F, Árnadóttir T, Hjartardóttir AR, Lárentínusdóttir M, Einarsson P, Schmidth P (2018) Reawakening of the Öræfajökull volcano in Iceland: deformation signals of stress triggers and intrusive activity. EGU Gen Assembly 20:15438Google Scholar
- Gíslason SR, Stefánsdóttir G, Pfeffer MA, Barsotti S, Jóhannsson Þ, Galeczka I, Bali E, Sigmarsson O, Stefánsson A, Keller NS, Sigurdsson Á, Bergsson B, Galle B, Jacobo VC, Arellano S, Aiuppa A, Jónasdóttir EB, Eiríksdóttir ES, Jakobsson S, Gudfinnsson GH, Halldórsson SA, Gunnarsson H, Haddadi B, Jónsdóttir I, Thordarson T, Riishuus M, Högnadóttir T, Dürig T, Pedersen GBM, Höskuldsson Á, Gudmundsson MT (2015) Environmental pressure from the 2014–15 eruption of Bárðarbunga volcano, Iceland. Geochem Perspect Lett 1:84–93CrossRefGoogle Scholar
- Grapenthin R, Sigmundsson F, Geirsson H, Árnadóttir T, Pinel V (2006) Icelandic rhythmics: annual modulation of land elevation and plate spreading by snow load. Geophys Res Lett 33: L24305, https://doi.org/10.1029/2006GL028081
- Grapenthin R, Siqi L, Ofeigsson B, Sigmundsson F, Vincent D, Hreinsdottir S, Parks M, Fridriksdottir H. (2018) Post-eruptive Deformation after the 2014-5 Bárdarbunga dike intrusion and rifting event, Iceland, EGU General Assembly 2018 20:15632Google Scholar
- Gudmundsson MT, Jónsdóttir K, Hooper A, Holohan EP, Halldórsson SA, Ófeigsson BG, Cesca S, Vogfjörd KS, Sigmundsson F, Högnadóttir T, Einarsson P, Sigmarsson O, Jarosch AH, Jónasson K, Magnússon E, Hreinsdóttir S, Bagnardi M, Parks MM, Hjörleifsdóttir V, Pálsson F, Walter TR, Schöpfer MPJ, Heimann S, Reynolds HI, Dumong S, Bali E, Gudfinnsson GH, Dahm T, Roberts MJ, Hensch M, Belart JMC, Spaans K, Jakobsson S, Gudmundsson GB, Fridriksdóttir HM, Drouin V, Dürig T, Adalgeirsdóttir G, Riishuus MS, Pederson GBM, Boeckel T, Oddsson B, Pfeffer MA, Barsotti S, Bergsson B, Donovan A, Burton MR, Aiuppa A (2016) Gradual caldera collapse at Bárdarbunga volcano, Iceland, regulated by lateral magma outflow. Science 353(6296):aaf8988CrossRefGoogle Scholar
- Hjaltadottir S, Vogfjord K (2010) Seismic evidence of magma transport in Eyjafjallajökull during 2009–2010, Abstract V21F-02 presented at 2010 fall meeting, AGU, December 13–17, San FranciscoGoogle Scholar
- Hjaltadóttir S, Vogfjörd KS, Slunga R (2009) Seismic Signs of Magma Pathways Through the Crust in the Eyjafjallajökull Volcano, South Iceland. Rep. VI 2009-13. Icelandic Meteorological Office, ReykjavíkGoogle Scholar
- Hreinsdóttir S, Einarsson P, Jónsson S, Sigmundsson F (1998) GPS geodetic measurements around Bárdarbunga volcano, central Iceland, in 1997, Report RH-13-98 Science Institute, University of Iceland, and Report 9802 Nord Volcanol Inst, p 41Google Scholar
- Hreinsdóttir S, Sigmundsson F, Roberts M, Árnadottir T, Ófeigsson B, Grapenthin R, Sturkell E, Villemin T, Bennett R, Geirsson H (2012) The 2010 Eyjafjallajökull and 2011 Grímsvötn eruptions: insights from GPS geodesy, geophysical research abstracts. EGU Gen Assembly 14:EGU2012-13577-1Google Scholar
- Hreinsdóttir S, Sigmundsson F, Roberts MJ, Grapenthin R, Arason P, Árnadóttir T, Hólmjárn J, Geirsson H, Bennett RA, Gudmundsson MT, Oddsson B, Ófeigsson BG, Villemin T, Jónsson T, Sturkell E, Höskuldsson Á, Larsen G, Thordarson T, Óladóttir BA (2014) Volcanic plume height correlated with magma pressure change at Grímsvötn volcano, Iceland. Nat Geosci 7:214–218. https://doi.org/10.1038/ngeo2044 ADSCrossRefGoogle Scholar
- Jóhannesson H (1982) Overview of the geology of the Snæfellsnes peninsula (in Icelandic). Arbok Ferdafelags Isl 151–172Google Scholar
- Jóhannesson H, Saemundsson K (2009a) Geological map of Iceland 1:600000, bedrock geology. Icelandic Institute of Natural History, ReykjavíkGoogle Scholar
- Jóhannesson H, Saemundsson K (2009b) Geological map of Iceland 1:600000, Tectonics. Icelandic Institute of Natural History, ReykjavíkGoogle Scholar
- Jónsdóttir K, Gudmundsson MT, Pfeffer M, Stefánsson A, Parks M, Lecocq T, Heidar Thrastarson R, Geirsson H, Barsotti S, Magnússon E, Pálsson F (2018) Reawakening of Öræfajökull volcano monitored using a multidisciplinary approach. EGU Gen Assembly 20:17969Google Scholar
- Jónsson ÓP (2013) GPS-measurements around the Eyjafjallajökull volcano since the end of the 2010 eruption (in Icelandic with English abstract), BSc. Thesis, University of IcelandGoogle Scholar
- Juncu D, Árnadóttir T, Geirsson H, Guðmundsson GB, Lund B, Gunnarsson G, Hooper A, Hreinsdóttir S, Michalczewska K (2018) Injection-induced surface deformation and seismicity at the Hellisheidi geothermal field, Iceland. J Volcanol Geotherm Res. https://doi.org/10.1016/j.jvolgeores.2018.03.019
- Kristinsson GH (2014) GPS Geodetic Measurements on Snæfellsnes Peninsula (in Icelandic). B.Sc. thesis, University of Iceland, Reykjavík, p 26Google Scholar
- Larsen G, Gudmundsson MT (2014) Volcanic system: bárdarbunga system. Pre-publication extract from the catalogue of Icelandic Volcanoes, published 25 Aug 2014Google Scholar
- Massonnet D, Sigmundsson F (2000) Remote sensing of volcano deformation by radar interferometry from various satellites. Remote Sens Active Volcanism 116:207–221Google Scholar
- Michalczewska K, Hreinsdóttir S, Dumont S, Sigmundsson F (2014) Crustal deformation on the Western Reykjanes Peninsula 2009 to 2014 mapped by InSAR and GPS measurements. Institute of Earth Sciences reportGoogle Scholar
- Michalczewska K, Hreinsdóttir S, Árnadóttir T, Hjaltadóttir S, Ágústsdóttir T, Gudmundsson MT, Geirsson H, Sigmundsson F, Gudmundsson G (2012) Inflation and deflation episodes in the Krísuvík volcanic system, Abstract, V33A-2843, AGU Fall MeetingGoogle Scholar
- Niemczyk O (1943) The splitting of Iceland – the German geological, geodetic and geophysical research expedition of 1938 (in German). Konrad Wittwer, Stuttgart, p 180Google Scholar
- Ófeigsson BG, Hreinsdóttir S, Sigmundsson F, Spaans K, Vogfjörd K, Hooper AJ, Sturkell EC, Roberts MJ, Gudmundsson GB, Einarsson P, Árnadóttir T, Jakobsdóttir S, Sigurdsson G, Gudmundsson MT, Geirsson H, Bennett RA (2011b) Increased volcanic unrest at Katla volcano, Iceland? Abstract C53E-2690. AGU Fall Meeting, San FranciscoGoogle Scholar
- Ófeigsson BG, Hreinsdóttir S, Sigmundsson F, Friðriksdóttir H, Parks M, Dumont S, Árnadóttir TH, Geirsson H, Hooper A, Roberts M, Bennett R, Sturkell E, Jonsson S, LaFemina P, Jonsson T, Steinthorsson S, Einarsson P, Drouin V (2015) Deformation derived from GPS geodesy associated with Bárðarbunga 2014 rifting event in Iceland. Geophys Res Abstr 17:EGU2015-7691-4Google Scholar
- Óskarsson BV (2009) The Sker ridge on Eyjafjallajökull, South Iceland: morphology and magma-ice interaction in an ice-confined silicic fissure eruption, MSc thesis, University of IcelandGoogle Scholar
- Parks M, Heimisson ER, Sigmundsson F, Hooper A, Vogfjörd K, Árnadóttir T, Ófeigsson B, Hreinsdóttir S, Hjartardóttir ÁR, Einarsson P, Gudmundsson MT, Högnadóttir T, Jónsdóttir K, Hensch M, Bagnardi M, Dumont S, Drouin V, Spaans K, Ólafsdóttir R (2017) Evolution of deformation and stress changes during the caldera collapse and dyking at Bárdarbunga, 2014–2015: Implication for triggering of seismicity at nearby Tungnafellsjökull volcano. Earth Planet Sci Lett 462:212–223ADSCrossRefGoogle Scholar
- Parks M, Sigmundsson F, Sigurðsson Ó, Hooper A, Hreinsdóttir S, Ófeigsson B, Michalczewska K (2018) Deformation due to geothermal exploitation at Reykjanes, Iceland. J Volcanol Geotherm Res. https://doi.org/10.1016/j.jvolgeores.2018.08.016
- Sæmundsson K (1978) Fissure swarms and central volcanoes of the neovolcanic zones of Iceland. Geol J Spec Issue 10:415–432Google Scholar
- Saunders AD, Fitton JG, Kerr AC, Norry MJ, Kent RW (1997) The North Atlantic Igneous Province. In: Mahoney JJ, Coffin MF (eds) Large igneous provinces, continental, oceanic, and planetary flood volcanism, Geophys. Monogr, vol 1000. American Geophysical Union, Washington, DC, pp 45–93Google Scholar
- Scheiber SE (2008) Geodetic investigation of Torfajökull volcano Iceland, MSc thesis, University of Witwatersrand JohannesburgGoogle Scholar
- Sigmundsson F (2006) Iceland geodynamics. Praxis Publishing Limited, ChichesterGoogle Scholar
- Sigmundsson F, Einarsson P, Rögnvaldsson ST, Foulger GR, Hodgkinsson KM, Thorbergsson G (1997b) The 1994–1995 seismicity and deformation at the Hengill triple junction, Ice land: triggering of earthquakes by minor magma injection in a zone of horizontal shear stress. J Geophys Res 102:15151–15161ADSCrossRefGoogle Scholar
- Sigmundsson F, Hreinsdóttir S, Hooper A, Árnadóttir T, Pedersen R, Roberts MJ, Óskarsson N, Auriac A, Decriem J, Einarsson P, Geirsson H, Hensch M, Ófeigsson B, Sturkell E, Sveinbjörnsson H, Feigl KL (2010) Intrusion triggering of the 2010 Eyjafjallajokull explosive eruption. Nature 468:426–430. https://doi.org/10.1038/nature09558 ADSCrossRefGoogle Scholar
- Sigmundsson F, Hreinsdóttir S, Sturkell E, Ófeigsson B, Einarsson P, Roberts M, Grapenthin R, Villemin T, Árnadóttir T, Geirsson H (2014) Grímsvötn volcano, Iceland, 1992–2014: constraints on magma flow in relation to eruptions in 1998, 2004 and 2011. Gephys ResAbstr 16:EGU2014-10561Google Scholar
- Sigmundsson F, Hooper A, Hreinsdóttir S, Vogförd KS, Ófeigsson BG, Heimisson ER, Dumont S, Parks M, Spaans K, Gudmundsson GB, Drouin V, Árnadóttir T, Jónsdóttir K, Gudmundsson MT, Högnadottir T, Fridriksdóttir HM, Hensch M, Einarsson P, Magnússon E, Samsonov S, Brandsdóttir B, White RS, Ágústsdóttir T, Greenfield T, Green RG, Hjartardóttir ÁR, Pedersen R, Bennett RA, Geirsson H, LaFemina P, Björnsson H, Pálsson F, Sturkell E, Bean CJ, Möllhoff M, Braiden AK, Eibl EPS (2015) Segmented lateral dyke growth in a rifting event at Bárdarbunga volcanic system, Iceland. Nature 517:191–195ADSCrossRefGoogle Scholar
- Sigmundsson F, Einarsson P, Hjartardóttir ÁR, Drouin V, Jónsdóttir K, Árnadóttir T, Geirsson H, Hreinsdóttir S, Li S, Ófeigsson BG (2018) Geodynamics of Iceland and the signatures of plate spreading. J Volcanol Geotherm Res. https://doi.org/10.1016/j.jvolgeores.2018.08.014
- Sigurdardóttir SS, Gudmundsson MT, Hreinsdóttir S (2015) Mapping of the Eldgjá lava flow on Mýrdalssandur with magnetic surveying. Jökull 65:61–71Google Scholar
- Spaans K (2011) Deformation at Katla volcano, Iceland, 2003–2009: disentangling surface displacements due to ice load reduction and magma movement using in SAR time series analysis, MSc Thesis, Delft University of TechnologyGoogle Scholar
- Sturkell E, Ágústsson K, Linde AT, Sacks SI, Einarsson P, Sigmundsson F, Geirsson H, Pedersen R, LaFemina PC, Ólafsson H (2013) New insights into volcanic activity from strain and other deformation data for the Hekla 2000 eruption. J Volcanol Geotherm Res 256:78–86. https://doi.org/10.1016/j.jvolgeores.2013.02.001 ADSCrossRefGoogle Scholar
- Tarasewicz J, Brandsdóttir B, White RS, Hensch M, Thorbjarnardóttir B (2012) Using microearthquakes to track repeated magma intrusions beneath the Eyjafjallajökull stratovolcano, Iceland. J Geophys Res 117. https://doi.org/10.1029/2011JB008751
- Thórarinsson S (1967) The eruptions of Hekla in historical times. In: Einarsson T, Kjartansson G, Thorarinsson S (eds) The eruption of Hekla 1947–1948. I. Soc Sci Islandica, Reykjavík, pp 1–170Google Scholar
- Thrastarson S (2013) Far-field GPS measurements around Eyjafjallajökull in 2010 (in Icelandic with English abstract), BSc thesis, University of IcelandGoogle Scholar
- Tryggvason E (1982) The NVI magnetoresistor tiltmeter, results of observations 1977–1981. Report, Nordic Volcanological Institute, 8203, p 44Google Scholar
- Tryggvason E (1995) Optical leveling tilt stations in the vicinity of Krafla and the Krafla fissure swarm – Observations 1976 to 1994. Report, Nordic Volcanological Institute 9505, ReykjavíkGoogle Scholar