One year of cyclic unrest in a hydrothermal field as a harbinger of a volcanic eruption

Abstract

One year of deformation and seismicity prior to a volcanic eruption in March 2021 at an oblique plate boundary in Iceland created a unique opportunity to study the interaction between upwelling magma and geothermal processes. We apply poroelastic modelling to explain satellite geodetic data showing three uplift and subsidence cycles at the Svartsengi geothermal field and use gravity data to constrain the density of intruded material. We use recordings on optical cable to generate a high-resolution earthquake catalogue and developed new waveform stacking and migration methods to detect and locate 39,500 earthquakes. The resulting model explains the geodetic, gravity, and seismic data by magmatic derived gas intruded into a horizontal sealed aquifer at 4 km depth in the roots of the geothermal field at the top of up-doming brittle-ductile boundary. The total injected volume is estimated 9.5·107m3 with optimal density of 840 kg/m3. Our results suggest upward migration of three packages of volcanic gas along the brittle-ductile boundary from a subcrustal magmatic source 8–10 km east of the geothermal field, with important implications for the dynamics leading to the eruption.

Evolving Shallow-conduit Container Affects the Lava Fountaining during the 2021 Fagradalsfjall Eruption, Iceland

“…However, before 2021 it had not erupted in more than 6000 years (Saemunds-son and Sigurgeirsson, 2013) and hence does not follow this episodic pattern of volcanism on the Peninsula. The 2021 eruption at Fagradalsfjall may be signalling the onset of a new eruption period on the Reykjanes Peninsula (Çubuk-Sabuncu et al, 2021;Flóvenz et al, 2022;Geirsson et al, 2021).…”

Stacking of Distributed Dynamic Strain Reveals Link Between Seismic Velocity Changes and the 2020 Unrest in Reykjanes

“…Depths range between several meters and 10 km. Events are particularly shallow beneath the geothermal field where the brittle‐ductile boundary domes up and deeper further away from it (Flovenz et al., 2022). We find no evidence of an initial occurrence of earthquakes further in the east followed by a subsequent westward progression over time, however, we cannot rule out that seismicity affects our measurements.…”

“…The tendency for velocity changes to be measured first in the east and then in the west is generally consistent with our observations. Flovenz et al (2022) propose a model for the observed deformation in Reykjanes, in which each inflation period is caused by the intrusion of magmatic fluids into an aquifer at approximately 4 km depth. The aquifer resembles the approximate shape and location of the geothermal field outlined in Figure 1.…”

“…Our study focuses on the Reykjanes peninsula (SW Iceland), where an intense unrest period took place in 2020 in the area of the Svartsengi geothermal field (Figure 1). The unrest period consisted of three episodes of crustal inflation between January and August of 2020 and preceded a series of eruptions (Flovenz et al., 2022). Each inflation episode was followed by crustal deflation and associated with a large number of small to moderate earthquakes ( M < 5).…”

Stacking of distributed dynamic strain reveals link between seismic velocity changes and the 2020 unrest in Reykjanes

“…Our study focuses on the Reykjanes peninsula (SW Iceland), where an intense unrest period took place in 2020 in the area of the Svartsengi geothermal field (Figure 1). The unrest period consisted of three episodes of crustal inflation between January and August of 2020 and preceded a series of eruptions (Flovenz et al, 2022). Each inflation episode was followed by crustal deflation and associated with a large number of small to moderate earthquakes (M<4.8).…”