On the Origin of Seismic Anisotropy in the Shallow Crust of the Northern Volcanic Zone, Iceland

Abstract

The Icelandic crust is a product of its unique tectonic setting, where the interaction of an ascending mantle plume and the Mid‐Atlantic Ridge has caused elevated mantle melting, with the melt accreted and cooled in the crust to form an oceanic plateau. We investigate the strength and orientation of seismic anisotropy in the upper crust of the Northern Volcanic Zone using local earthquake shear‐wave splitting, with a view to understanding how the contemporary stress field may influence sub‐wavelength structure and processes. This is achieved using a data set comprising > $ > $50,000 earthquakes located in the top 10 km of the crust, recorded by up to 70 stations over a 9 year period. We find that anisotropy is largely confined to the top 3–4 km of the crust, with an average delay time of 0.10 ± 0.05 s, and an average orientation of the fast axis of anisotropy of N014°E ± 27°, which is perpendicular to the spreading direction of the Eurasian and North American plates (N106°E). These results are consistent with the presence of rift‐parallel cracks that gradually close with depth, the preferential opening of which is controlled by the regional stress field. Lateral variations in the strength of shear wave anisotropy (SWA) reveal that regions with the highest concentrations of earthquakes have the highest SWA values (∼10%), which reflects the presence of significant brittle deformation. Disruption of the orientation of the fast axis of anisotropy around Askja volcano can be related to local stress changes caused by underlying magmatic processes.

Evidence of a Shallow Magma Reservoir Beneath Askja Caldera, Iceland, From Body Wave Tomography

“…Anomaly III is mainly concentrated in the highly fractured zones, between Askja and Herðubreið in this case, which is similar to observations in the San Andreas fault and Eastern California Shear Zone (Berg et al., 2021; Fang et al., 2018; Lin et al., 2007). In particular, Vs is found to be sensitive to highly fractured rocks (Catchings et al., 2020), but this diminishes with depth as cracks begin to close, which is also supported by studies of seismic anisotropy (Bacon et al., 2022).…”

“…Grímsvötn, the most active volcano in Iceland, lies beneath the Vatnajokull icecap. It features a low-velocity anomaly, which is surrounded by highvelocity anomalies in the shallow crust (Alfaro et al, 2007). Geodetic modeling suggests a magma body exists at ∼1 km bsl, and Grímsvötn is similar to Askja and Krafla in that the pressure change of these upper-crustal magma bodies drives the alternation between deflation and inflation (Hreinsdóttir et al, 2014).…”

“…Furthermore, the orientation of the fast axis of anisotropy inside the caldera appears to be scattered in comparison to those outside the caldera, which tend to be rift parallel. Stress modeling (Bacon et al., 2022) supports this disruption being caused by a shallow contracting magma body.…”

“…A recent seismic anisotropy study that exploited local earthquakes (Bacon et al., 2022) detected the presence of shallow fracturing in the caldera and its surrounding regions, which is also indicated by the dense microseismicity (Greenfield et al., 2020; Winder, 2021). Furthermore, the orientation of the fast axis of anisotropy inside the caldera appears to be scattered in comparison to those outside the caldera, which tend to be rift parallel.…”

“…The existence of a shallow magma body beneath Askja caldera has been supported by studies from multiple disciplines, including but not limited to geodetic modeling (de Zeeuw‐van Dalfsen et al., 2012); ambient noise tomography (Volk, 2021); observations of higher attenuation associated with seismic rays passing directly beneath the caldera compared to those that do not (Greenfield et al., 2016); local earthquake shear wave splitting analysis revealing scattered fast axis orientations of anisotropy beneath the caldera (Bacon et al., 2022); and a petrological study (Sigurdsson & Sparks, 1981). In the case of the ambient noise study, both horizontal and depth resolution were limited by the period range used, resulting in a somewhat nebulous low‐velocity anomaly that has no clear spatial association with the current uplift.…”