The 7.8Mw Kaikoura earthquake that occurred in New Zealand on November 14th 2016 has no similar analogue on earth since the beginning of instrumental measurements. It is a dual epicentre, two minute-long rupture that involved slip on >20 individual faults rooted at different crustal depths. In principle, this earthquake links deep thrust faulting with upper-crust faulting through an ongoing composite rupture. Interestingly, our research over the last 5 years has highlighted the importance of such an interaction on subduction margins globally and how crucial this interaction may be for the estimation of the repeat times of mega-earthquakes. Here we have a unique opportunity to elaborate this concept further as this is the first known earthquake to record such fault interaction through a single rupture. Therefore, with this project we seek to explore how thrust-faulting and upper-plate faulting interact to accommodate large-magnitude earthquakes near subduction terminations globally. During fieldwork in the broader Kaikoura region we mapped ca. 55 km of coastline to record the coseismic deformation using tape and Real Time Kinematics GPS measurements of the uplifted intertidal rocky shore biota. We performed drone-mapping of key uplifted shorelines. We analyzing instrumentally recorded InSAR, LiDAR and GPS measurements. We also studied and sampled for OSL dating the Pleistocene marine terraces along the Kaikoura coastline to capture long-term deformation.

By focusing on the Kaikoura earthquake we will:

• Understand the role of splay-thrust faults in subduction seismogenesis: There are instances where mega-earthquakes on the plate-interface trigger rupture on upper-plate forearc splay faults as, for example, in the 2010 Maule earthquake (Moreno et al., 2010). By contrast, McCaffrey and Goldfinger (1995) argue that in Cascadia, upper-plate forearc faults and the related non-elastic deformation exert an important role on the lack of thrust mega-earthquakes by inhibiting build-up of elastic strain energy and/or slip propagation. Interestingly, upper-plate faulting has been observed both prior (Schurr et al., 2014; Hayes et al., 2014) and after (Plafker, 1967; Farias et al., 2011) great subduction earthquakes at several active plate-boundaries worldwide. Hence, although their temporal interaction is poorly constrained, it is clear that splay thrust-faults have a direct impact on the kinematics and stress state of the plate-interface, affecting the accumulation (and release) of strain on the interface and in the upper-plate. By identifying the geometry and the short- and long-term kinematics of the large thrust fault involved in the Kaikoura earthquake, we will be in position to know whether its motion has persisted through geological timescales and at which rates. In doing so, we will be able to derive an estimate of the recurrence interval on this unknown, up to date fault, attribute which is vital for the assessment of the seismic hazard.

• Understand the interaction of subduction-thrust faulting and shallower upper-crust faulting along active plate-boundaries: It is of vital importance to better understand the temporal and spatial interaction between fault-systems rooted at various depths of a subduction margin. Having already established robust data on the long-term kinematics of the strike-slip faulting at the South Island of New Zealand, and constraining the long-term kinematics of the deep-thrust events (see above) will be able to identify their temporal/spatial relation over thousand yearlong timescales. This will be a unique case-study as, up-to-date, other subduction systems offer long-term records on either thrust faulting (e.g., the Hellenic), or plate-interface faulting (e.g., Chile) or upper-plate faulting (e.g., New Zealand) alone. In this case we will able to model, using natural empirical earthquakes from thousand yearlong timescales and co-seismic measurements from this extraordinary earthquake, the earthquake interaction on various synchronously operating fault systems of a subduction margin.

• Modelling complex rupture and aftershocks patterns: First seismological analysis of the main rupture mechanism, its source time function and aftershocks sequence, support a scenario of complex faulting, with at least two dominant mechanisms active during the main shock. These conditions offer a unique chance to investigate a complex multiple source rupture process in combination with its surface expression. We plan a seismological analysis of the complex rupture combined with a stress based modelling of the aftershock distribution. With no need of dedicated installation, since open broadband seismic, accelerometer and GPS data are provided by partners in New Zealand. The seismological research would strongly benefit from field mapping of surface uplift, which will help to constrain kinematic and dynamic source models.

Team members: Vasiliki Mouslopoulou, Onno Oncken, Sofia Kufner all sec. 4.1), Simone Cesca, Torsten Dahm (both sec. 2.1), Andy Nicol (Canterbury University, NZ), John Begg (GNS Science, NZ)

Funding: HART GFZ

Publications: Cesca, S., Zhang, Y., Mouslopoulou, V., Wang, R., Saul, J., Savage, M., Heimann, S., Kufner, S.-K., Oncken, O., Dahm, T., 2017. Complex rupture process of the Mw 7.8, 2016, Kaikoura earthquake, New Zealand, and its aftershock sequence. Earth and Planetary Science Letters, 478, 110-120, https://doi.org/10.1016/j.epsl.2017.08.024.

Spectacular slip during the earthquake in New Zealand (Photo: John Begg, GNS Science, NZ).

Survey of the uplifted coastline after the Kaikoura Earthquake in New Zealand (Photo: Vasiliki Mouslopoulou, GFZ).

Survey of the inaccessible section of the fault rupture using boat (Photo: Dick Beetham, GHD, NZ).