Maria Liukis

The forecasting skill of physics-based seismicity models during the 2010-2012 Canterbury, New Zealand, earthquake sequence

1 The static Coulomb stress hypothesis is a widely known physical mechanism for 2 earthquake triggering, and thus a prime candidate for physics-based Operational Earth3 quake Forecasting (OEF). However, the forecast skill of Coulomb-based seismicity mod4 els remains controversial, especially in comparison to empirical statistical models. A 5 previous evaluation by the Collaboratory for the Study of Earthquake Predictabil6 ity (CSEP) concluded that a suite of Coulomb-based seismicity models were less in7 formative than empirical models during the aftershock sequence of the 1992 Mw7.3 8 Landers, California, earthquake. Recently, a new generation of Coulomb-based and 9 Coulomb/statistical hybrid models were developed that account better for uncertainties 10 and secondary stress sources. Here, we report on the performance of this new suite of 11 models in comparison to empirical Epidemic Type Aftershock Sequences (ETAS) mod12 els during the 2010-2012 Canterbury, New Zealand, earthquake sequence. Comprising 13 the 2010 M7.1 Darfield earthquake and three subsequent M ≥ 5.9 shocks (including 14 the February 2011 Christchurch earthquake), this sequence provides a wealth of data 15 (394 M ≥ 3.95 shocks). We assessed models over multiple forecast horizons (1-day, 16 1-month and 1-year, updated after M ≥ 5.9 shocks). The results demonstrate substan17 tial improvements in the Coulomb-based models. Purely physics-based models have a 18 performance comparable to the ETAS model, and the two Coulomb/statistical hybrids 19 perform better or as well as the corresponding statistical model. On the other hand, 20 an ETAS model with anisotropic (fault-based) aftershock zones is just as informative. 21 These results provide encouraging evidence for the predictive power of Coulomb-based 22 models. To assist with model development, we identify discrepancies between forecasts 23 and observations. 24