Fault models are quickly produced and iteratively improved over weeks to years following a major earthquake, to characterise the dynamics of rupture, evaluate the role of stress transfer, and contribute to earthquake forecasting. We model Coulomb stress transfer (Delta CFS) between the largest foreshock (Mw 5.4; 1 year prior to first mainshock) and three Mw 6.1 to 6.5 earthquakes that occurred in a 12-hour period on January 22, 1988 in central Australia (Tennant Creek earthquake sequence) to investigate the role of static stress transfer in earthquake triggering relative to progressive source model development. The effects of fault model variance are studied using Delta CFS modelling of five different fault source model sequences (27 total models) using different inputs from seismic and geospatial data. Some initial models do not yield positive Delta CFS changes proximal to hypocentres but in all models, preceding earthquakes generate positive Delta CFS (>= 0.1 bar) on >= a 10 to 30% of the forthcoming receiver fault rupture areas. The most refined and data-integrative model reveals Delta CFS >= + 0.7 to +13 bars within 2 km of impending hypocentres and large (>= 30 to 99%) areas of positive ACES. When compared to global compilations of threshold Delta CFS prior to impending ruptures (average = 3.71 bar, median = 1 bar), this suggests that Coulomb stress change theory adequately explains the Tennant Creek rupture sequence. In the most-refined model, earthquake inter-event times decrease as Delta CES increases, suggesting that higher stress magnitudes may have more rapidly (within hours) triggered successive events, thus accounting for some temporal aspects of this sequence. Delta CFS analyses provide a useful framework for understanding the spatiotemporal aspects of some intraplate earthquakes. The progressive refinement of source models using emergent data may reduce epistemic uncertainties in the role of stress transfer that result from different model inputs, approaches, and results.

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

Aftershock sequences are an ideal testing ground for operational earthquake forecasting models as they contain relatively large numbers of earthquakes clustered in time and space. To date, most successful forecast models have been statistical, building on empirical observations of aftershock decay with time and earthquake size frequency distributions. Another approach is to include Coulomb stress changes from the mainshock which influence the spatial location of the aftershocks although these models have generally not performed as well as the statistical ones. Here we develop a new hybrid Epidemic-Type Aftershock Sequence (ETAS)/Coulomb model which attempts to overcome the limitations of its predecessors by redistributing forecast rate from negatively to positively stressed regions based on observations in the model learning period of the percentage of events occurring in those positively stressed regions. We test this model against the 1992 Landers aftershock sequence using three different ETAS kernels and five different models for slip in the Landers earthquake. We also consider two variations of rate redistribution, one based on a fixed value and the other variable depending on the percentage of aftershocks observed in positively stressed Coulomb regions during the learning period. We find that the latter model performs at least as well as ETAS on its own in all tests and better than ETAS in 14 of 15 tests in which we forecast successive 24-hr periods. Our results suggest that including Coulomb stress changes can improve operational earthquake forecasting models.

We analyse the influence of the solid Earth tides and ocean loading on the occurrence time of Southern California earthquakes. For each earthquake, we calculate tidal Coulomb failure stress and stress rate on a fault plane that is assumed to be controlled by the orientation of the adjacent fault. To reduce bias when selecting data for testing the tide-earthquake relationship, we create four earthquake catalogues containing events within 1, 1.5, 2.5 and 5 km of nearest faults. We investigate the difference in seismicity rates at times of positive and negative tidal stresses/ stress rates given three different cases. We consider seismicity rates during times of positive versus negative stress and stress rate, as well as 2 and 3 hr surrounding the local tidal stress extremes. We find that tidal influence on earthquake occurrence is found to be statistically non-random only in close proximity to tidal extremes meaning that magnitude of tidal stress plays an important role in tidal triggering. A non-random tidal signal is observed for the reverse events. Along with a significant increase in earthquake rates around tidal Coulomb stress maxima, the strength of tidal correlation is found to be closely related to the amplitude of the peak tidal Coulomb stress (tau(p)). The most effective tidal triggering is found for tau(p) >= 1 kPa, which is much smaller than thresholds suggested for static and dynamic triggering of aftershocks.

We investigate the postseismic viscoelastic flow in the lower crust and upper mantle due to the 19th and 20th century large earthquakes in eastern Turkey. Three possible rheological models are used in the viscoelastic postseismic deformation analysis to assess the extent to which these events influence the velocity fields at GPS sites in the region. Our models show that the postseismic signal currently contributes to the observed deformation in the eastern part of the North Anatolian fault and northern and middle parts of the East Anatolian Fault Zone, primarily due to the long-lasting effect of the Ms 7.9 1939 earthquake. None of the postseismic displacement generated by the Ms 7.5 1822 earthquake, which is the earliest and the second largest event in the calculations, exceeds observed error range at the GPS stations. Our results demonstrate that a postseismic signal can be identified in the region and could contribute up to 3–25% of the observed GPS measurements.

Risk communication, where scientists inform policy-makers or the populace of the probability and magnitude of possible disasters, is essential to disaster management – enabling people to make better decisions regarding preventative steps, evacuations, etc. Psychological research, however, has identified multiple biases that can affect people’s interpretation of probabilities and thus risk. For example, availability (Tversky & Kahneman, 1973) is known to confound probability estimates while the descriptionexperience gap (D-E Gap) (Hertwig & Erev, 2009) shows low probability events being over-weighted when described and under-weighted when learnt from laboratory tasks. This paper examines how probability descriptions interact with real world experience of events. Responses from 294 participants across 8 conditions showed that people’s responses, given the same described probabilities and consequences, were altered by their familiarity with the disaster (bushfire vs earthquake) and its salience to them personally. The implications of this for risk communication are discussed.

The extent to which interseismic coupling controls the slip distribution of large megathrust earthquakes is unclear, with some authors proposing that it is the primary control and others suggesting that stress changes from previous earthquakes are of first‐order importance. Here, we develop a detailed stress history of the Sunda megathrust, modified by coupling, and compare the correlation between slip and stress with that of slip versus coupling. We find that the slip distributions of recent earthquakes are more consistent with the stress field than with the coupling distributions but observe that in places, the stress pattern is strongly dependent on poorly constrained values of slip in historical earthquakes. We also find that of the 13 earthquakes in our study for which we have hypocentral locations, only two appear to have nucleated in areas of negative stress, and these locations correspond to large uncertainties in the slip distribution of pre‐instrumental events.

Forecasting the spatial and temporal distribution of aftershocks is of great importance to earthquake scientists, civil protection authorities and the general public as these events cause disproportionate damage and consternation relative to their size. At present, there are two main approaches to such forecasts-purely statistical methods based on observations of the initial portions of aftershock sequences and a physics-based approach based on Coulomb stress changes caused by the main shock. Here we develop a new method which combines the spatial constraints from the Coulomb model with the statistical power of the STEP (short-term earthquake probability) approach. We test this pseudo prospectively and retrospectively on the Canterbury sequence against the STEP model and a Coulomb rate-state method, using data from the first 10 d following each main event to forecast the rate of M >= 4 events in the following 100 d. We find that in retrospective tests the new model outperforms STEP for two events in the sequence but this is not the case for pseudo-prospective tests. Further, the Coulomb rate-state approach never performs better than STEP. Our results suggest that incorporating the physical constraints from Coulomb stress changes can increase the forecasting power of statistical models and clearly show the importance of good data quality if prospective forecasts are to be implemented in practice.

The Canterbury earthquake sequence, which includes the devastating Christchurch event of 2011 February, has to date led to losses of around 40 billion NZ dollars. The location and severity of the earthquakes was a surprise to most inhabitants as the seismic hazard model was dominated by an expected Mw > 8 earthquake on the Alpine fault and an Mw 7.5 earthquake on the Porters Pass fault, 150 and 80 km to the west of Christchurch. The sequence to date has included an Mw = 7.1 earthquake and 3 Mw ≥ 5.9 events which migrated from west to east. Here we investigate whether the later events are consistent with stress triggering and whether a simple stress map produced shortly after the first earthquake would have accurately indicated the regions where the subsequent activity occurred. We find that 100 per cent of M > 5.5 earthquakes occurred in positive stress areas computed using a slip model for the first event that was available within 10 d of its occurrence. We further find that the stress changes at the starting points of major slip patches of post-Darfield main events are consistent with triggering although this is not always true at the hypocentral locations. Our results suggest that Coulomb stress changes contributed to the evolution of the Canterbury sequence and we note additional areas of increased stress in the Christchurch region and on the Porters Pass fault.