Representation of the 2016 Mw 7.8 Kaikoura Earthquake by Orthogonal Moment Tensor Fields

Date: 4/25/2019

Time: 11:00 AM

Room: Cascade I

The Mw 7.8 Kaikoura, New Zealand, earthquake of 2016 is one of the most complex earthquakes ever recorded. It ruptured more than twenty crustal faults with different strikes, dips, and rakes, and the subduction interface beneath New Zealand. We use the stress-glut representation theorem of Jordan & Juarez (GJI, 2018) to analyse finite fault models of the Kaikoura earthquake and quantify the complexity of the source. The representation theorem decomposes the stress-glut density into a set of up to six orthogonal moment tensor fields of increasing degree. These fields are ordered according to their first non-zero polynomial moments, which can be uniquely factored into a unit moment tensor (source mechanism) and a shift-invariant multipole tensor. The centroid moment tensor (CMT) is the monopole (0th-degree) term in the tensor multipole expansion of the general point source defined by the representation theorem. We decompose source models that include the Hikurangi subduction interface and compare them with models that do not. We estimate the source mechanisms and multipole tensors associated with the 0th- and higher-degree moment tensor fields. The 0th-degree source mechanisms of the models with subducting slab are in good agreement with the Global-CMT solution. The higher-degree fields quantify the complexity of the source that contributes to the seismic radiation not excited by the 0th-degree field. The characteristics of the source given by the magnitude and orientation of the dipole and quadrupole tensors are consistent among the models. We find that the Aki seismic moment represents 0.67-0.82 of the total moment (M_T), depending on the model. The stress-glut representation theorem allows to compute the fractions of M_T released by the higher-degree fields; their sum accounts 0.25-0.45 of M_T. Our results are consistent with the analytical calculations of Jordan & Juarez (2018) using a stochastic parametrization of source complexity, and they suggest that it may be possible to estimate low degree multipoles directly from seismic data.

Presenting Author: Alan Juarez

Authors