Seismicity and deformation associated with ice-shelf rift propagation

• Published in: Journal of Glaciology
• Main Authors: Bassis, JN, Fricker, HA, Coleman, R, Bock, Y, Behrens, J, Darnell, D, Okal, M, Minster, JB
• Format: Article in Journal/Newspaper
• Language: English
• Published: International Glaciological Society 2007
• Subjects: 260115 Glaciology; Amery; Amery Ice Shelf; East Antarctica; Antarc*; Antarctica; Ice Shelf; Journal of Glaciology
• Online Access: https://eprints.utas.edu.au/3876/; https://eprints.utas.edu.au/3876/1/Basis_et_al_2007.pdf; https://doi.org/10.3189/002214307784409207

Previous observations have shown that rift propagation on the Amery Ice Shelf (AIS), East Antarctica, is episodic, occurring in bursts of several hours with typical recurrence times of several weeks. Propagation events were deduced from seismic swarms (detected with seismometers) concurrent with rapid rift widening (detected with GPS receivers). In this study, we extend these results by deploying seismometers and GPS receivers in a dense network around the tip of a propagating rift on the AIS over three field seasons (2002/03, 2004/05 and 2005/06). The pattern of seismic event locations shows that icequakes cluster along the rift axis, extending several kilometers back from where the rift tip was visible in the field. Patterns of icequake event locations also appear aligned with the ice-shelf flow direction, along transverse-to-rift crevasses. However, we found some key differences in the seismicity between field seasons. Both the number of swarms and the number of events within each swarm decreased during the final field season. The timing of the slowdown closely corresponds to the rift tip entering a suture zone, formed where two ice streams merge upstream. Beneath the suture zone lies a thick band of marine ice. We propose two hypotheses for the observed slowdown: (1) defects within the ice in the suture zone cause a reduction in stress concentration ahead of the rift tip; (2) increased marine ice thickness in the rift path slows propagation. We show that the size–frequency distribution of icequakes approximately follows a power law, similar to the well-known Gutenberg–Richter law for earthquakes. However, large icequakes are not preceded by foreshocks nor are they followed by aftershocks. Thus rift-related seismicity differs from the classic foreshock and aftershock distribution that is characteristic of large earth quakes.