Insights Into the Stress Field Around Bárarbunga Volcano From the 2014/2015 Holuhraun Rifting Event

Abstract

The two weeklong rifting event at Bárðarbunga volcano in 2014 led to the Holuhraun eruption, which produced 1.5 km3 of lava and was the largest in Iceland in over 200 years. Predicting when and where an intrusion will lead to eruption requires detailed knowledge of the underlying stress field. Previous studies have explained the dike propagation path with a model that includes a tectonically induced stress field set up by a uniform amount of plate spreading across a straight rift axis. Here we test this hypothesis by modeling the tractions acting on the dike walls, constrained by data from Global Navigation Satellite System and Interferometric Synthetic Aperture Radar. Our results show that the majority of the opening and shearing in the final two dike segments is due to stresses built up by plate spreading since the last eruption at Holuhraun, as expected, but that the tectonically induced stress magnitude must be much lower to explain the movement of the dike walls further south. This result implies that most of the tectonically induced stress beneath the ice cap has been released, presumably due to intrusions associated with the Bárðarbunga volcanic system and the nearby Grímsvötn volcanic system, which have not been detected due to their subglacial nature. Modeling of the 2014 Bárðarbunga rifting event therefore not only yields insights into the event but also provides a window into undetected volcanic activity in the past.

Physically Consistent Modeling of Dike‐Induced Deformation and Seismicity: Application to the 2014 Bárðarbunga Dike, Iceland

“…Ruch et al () showed that a small amount of strike slip occurred on faults parallel to the dike and other deformation studies have also suggested that some slip occurred on the dike plane (Sigmundsson et al, ; Spaans & Hooper, ). Here we neglect this for two reasons.…”

“…The time span of the interferograms varies considerably with the later acquisition times ranging from 26 August to 20 September, 2014. The length of the dike likely did not change after 26 August (Sigmundsson et al, ; Spaans & Hooper, ), although, the dike pressure and the chamber volume were still evolving. is inherently time dependent because the acquisition times of the interferograms are variable.…”

“…The rows and columns of G corresponding to elements outside the periphery of the dike, including the tip cavity, are removed before the matrix is inverted to solve for the vector of opening b. Ruch et al (2016) showed that a small amount of strike slip occurred on faults parallel to the dike and other deformation studies have also suggested that some slip occurred on the dike plane (Sigmundsson et al, 2015;Spaans & Hooper, 2018). Here we neglect this for two reasons.…”

Evolution of a lateral dike intrusion revealed by relatively-relocated dike-induced earthquakes: The 2014–15 Bárðarbunga–Holuhraun rifting event, Iceland

“…Surface widening of ∼ 4.5 m was observed directly above the dike (Ruch et al, 2016), with up to ∼ 1 m spreading between GPS sites ∼ 10 km to the east and west of the dike (Sigmundsson et al, 2015). Geodetic modeling (Sigmundsson et al, 2015;Spaans and Hooper, 2018) suggests the dike was near-vertical and extended almost to the surface along its length, with variable opening on the order of several meters.…”

“…The intrusion was extensively monitored and there exist many studies -e.g. geodetic inversions (Sigmundsson et al, 2015;Spaans and Hooper, 2018), studies of ice cauldron, surface fracturing and graben formation (Reynolds et al, 2017;Hjartardóttir et al, 2016;Ruch et al, 2016;Rossi et al, 2016) -that can aid our interpretation and understanding of the overall process. The seismicity was recorded in such detail that we are also able to begin to address questions about the precise relationship between the dike and seismicity, with direct implications for hazard monitoring.…”