Seismicity and Structure of Subducting Slabs

Subduction zones host the majority of the Earth's seismic moment release. Whilst much of this is the result of motion between plates along the plate interface, significant seismicity also occurs within subducting plates, as they deform internally. Although they rarely reach the size or frequency of earthquakes associated with the subduction megathrust, these earthquakes, often located deep beneath cities landwards of major subduction zones, have the potential to be devastating, as seen in the magnitude 7.9 1970 Ancash (Peru - the deadliest earthquake in South American history), the 2001 magnitude 7.7 El Salvador earthquake, and, more recently, the magnitude 7.1 2017 Puebla/Mexico City earthquake (e.g., Melgar et al., 2018). In many cases, the occurrence of such earthquakes remains a surprise, with the capacity of the downgoing plate to host such large-magnitude earthquakes uncertain in many regions.

These earthquakes arise from the combination of the stresses derived from large-scale plate-driving forces, localised stresses arising from changes in slab geometry (e.g., changes in slab dip, slab tears, etc), and rheological factors relating to the structure and evolution of the subducting plate as it descends into the Earth's interior (Hacker et al., 2003; Abers et al., 2013). Critical to understanding the distribution of such earthquakes is our ability to accurately map out their location within the plate, in particular with respect to each other, and to the surface of the subducting plate (e.g., Abers et al., 2013). This project will initially focus on improving earthquake catalogues for a number of regional case studies (starting with South America), accurately mapping out the seismogenic structure of the subducting oceanic plate, the location and mode of failure of active faults within the plate, and how the distribution of stress and strain vary within the plate. This will principally be done through the detailed analysis and modelling of global seismic data, incorporating locally-acquired data where available and useful. Many of the techniques required have been previously developed, but the student will tailor existing approaches to the requirements of the study and datasets available, and will be involved in the development of new approaches to the assessment of the seismic data as required.

The project will also aim to answer the question of how these earthquakes relate to the geodynamic setting of the slab, and its structure and rheological evolution, through the combination of seismological observations and geodynamic modelling (Hacker et al., 2003). Improved controls on the location of slab seismicity will allow us to develop our understanding of how these earthquakes relate to the thermal structure of the slab as it starts to heat up, and the mineralogical phase transitions that the slab undergoes during subduction, and the role that these processes may play in localising or triggering seismicity.

A final aim will be to understand the relationship between the largest earthquakes occurring in such settings (M7-8), and the background seismicity in their vicinity. In particular, addressing the issue of how these earthquakes activate such large sections of the seismogenic slab, and whether their spatial occurrence can be predicted, and incorporated into seismic hazard models. Understanding the rupture extent of these earthquakes in comparison to the seismogenic structure of the slab, will allow us to address the question of what these earthquakes can tell us about both the geodynamics of the slab (Do such earthquake rupture the full extent of the seismogenic slab? Do they rupture through both double seismic zones?), and may also throw light on their causative mechanism, through comparison to smaller seismicity (e.g., Craig et al., 2014). This aspect of the project may involve more complex seismological modelling of these earthquakes, and consideration of their aftershock sequences.