Seismic hazard models, commonly produced through probabilistic seismic hazard analysis, are used to establish earthquake loading requirements for the built environment. However, there is considerable uncertainty in developing seismic hazard models, which require assumptions on seismicity rates and ground-motion models (GMMs) based on the best evidence available to hazard analysts. This paper explores several area-based tests of long-term seismic hazard forecasts for the Australian continent. ShakeMaps are calculated for all earthquakes of MW 4.25 and greater within approximately 200 km of the Australian coastline using the observed seismicity in the past 50 years (1970-2019). A “composite ShakeMap” is generated that extracts the maximum peak ground acceleration “observed” in this 50-year period for any site within the continent. The fractional exceedance area of this composite map is compared with four generations of Australian seismic hazard maps for a 10% probability of exceedance in 50 years (~1/500 annual exceedance probability) developed since 1990. In general, all these seismic hazard models appear to be conservative relative to the observed ground motions that are estimated to have occurred in the last 50 years. To explore aspects of possible prejudice in this study, the variability in ground-motion exceedance was explored using the Next Generation Attenuation-East GMMs developed for the central and eastern United States. The sensitivity of these results is also tested with the interjection of a rare scenario earthquake with an expected regional recurrence of approximately 5,000 - 10,000 years. While these analyses do not provide a robust assessment of the performance of the candidate seismic hazard for any given location, they do provide—to the first order—a guide to the performance of the respective maps at a continental scale. This paper was presented at the Australian Earthquake Engineering Society 2021 Virtual Conference, Nov 25 – 26.

We digitize surface rupture maps and compile observational data from 67 publications on ten of eleven historical, surface-rupturing earthquakes in Australia in order to analyze the prevailing characteristics of surface ruptures and other environmental effects in this crystalline basement-dominated intraplate environment. The studied earthquakes occurred between 1968 and 2018, and range in moment magnitude (Mw) from 4.7 to 6.6. All earthquakes involved co-seismic reverse faulting (with varying amounts of strike-slip) on single or multiple (1–6) discrete faults of ≥ 1 km length that are distinguished by orientation and kinematic criteria. Nine of ten earthquakes have surface-rupturing fault orientations that align with prevailing linear anomalies in geophysical (gravity and magnetic) data and bedrock structure (foliations and/or quartz veins and/or intrusive boundaries and/or pre-existing faults), indicating strong control of inherited crustal structure on contemporary faulting. Rupture kinematics are consistent with horizontal shortening driven by regional trajectories of horizontal compressive stress. The lack of precision in seismological data prohibits the assessment of whether surface ruptures project to hypocentral locations via contiguous, planar principal slip zones or whether rupture segmentation occurs between seismogenic depths and the surface. Rupture centroids of 1–4 km in depth indicate predominantly shallow seismic moment release. No studied earthquakes have unambiguous geological evidence for preceding surface-rupturing earthquakes on the same faults and five earthquakes contain evidence of absence of preceding ruptures since the late Pleistocene, collectively highlighting the challenge of using mapped active faults to predict future seismic hazards. Estimated maximum fault slip rates are 0.2–9.1 m Myr−1 with at least one order of uncertainty. New estimates for rupture length, fault dip, and coseismic net slip can be used to improve future iterations of earthquake magnitude—source size—displacement scaling equations. Observed environmental effects include primary surface rupture, secondary fracture/cracks, fissures, rock falls, ground-water anomalies, vegetation damage, sand-blows/liquefaction, displaced rock fragments, and holes from collapsible soil failure, at maximum estimated epicentral distances ranging from 0 to ~250 km. ESI-07 intensity-scale estimates range by ± 3 classes in each earthquake, depending on the effect considered. Comparing Mw-ESI relationships across geologically diverse environments is a fruitful avenue for future research.

Canada's 6th Generation seismic hazard model has been developed to generate seismic design values for the 2020 National Building Code of Canada (NBCC2020). The model retains most of the seismic source model from the 5th Generation, but updates the earthquake sources for the deep inslab earthquakes under the Straits of Georgia and adds the Leech River - Devil’s Mountain fault near Victoria. The rates of magnitude ~9 Cascadia earthquakes are also increased to match new paleoseismic information. Two major changes in the ground motion model (GMM) are A) replacement of most of the three-branch representative suite used in 2015 by suites of weighted GMMs, and B) use and adaptation of various GMMs to directly calculate hazard on various site classes with representative Vs30 values, rather than providing hazard values on a reference Class C site and applying F(T) factors as in 2015. Computations are now also being performed with the OpenQuake engine, which has been validated through the replication of the 5th Generation results. Seismic design values (on various Soil Classes) for PGA, and for Sa(T) for T = 0.2, 0.5, 1.0, 2.0, 5.0, and 10.0 s are proposed for NBCC2020 mean ground shaking at the 2% in 50-year probability level. The paper discusses chiefly the change in Site Class C values relative to 2015 in terms of the changes in the seismic source model and the GMMs, but the changes in hazard at other site classes that arise from application of the direct-calculation approach are also illustrated.

Because all modern ground motion prediction equations (GMPEs) are now calibrated to the moment magnitude scale MW, it is essential that earthquake rates are also expressed in terms of moment magnitudes for probabilistic seismic hazard analyses. However, MW is not routinely estimated for earthquakes in Australia because of the low-to-moderate level of seismicity, coupled with the relatively small number of seismic recording stations. As a result, the Australian seismic catalogue has magnitude measures mainly based on local magnitudes, ML. To homogenise the earthquake catalogue based on a uniform MW, a “reference catalogue” that includes earthquakes with available MW estimates was compiled. This catalogue consists of 240 earthquakes with original MW values between 2.0 and 6.58. This reference catalogue served as the basis for the development of magnitude conversion equations between MW and other magnitude scales: ML, body-wave magnitude mb, and surface-wave magnitude MS. The conversions were evaluated using general orthogonal regression (GOR), which accounts for measurement errors in the x and y variables, and provides a unique solution that can be used interchangeably between magnitude types. The impact of the derived magnitude conversion equations on seismic hazard is explored by generating synthetic earthquake catalogues and computing seismic hazard level at an arbitrary site. The results indicate that we may expect up to 20-40% reduction in PGA hazard, depending on the selection and application process of the magnitude conversion equations. Abstract submitted to and presented at the 2017 Australian Earthquake Engineering Society (AEES) Conference

Modern probabilistic seismic hazard assessments rely on earthquake catalogs consistently expressed in terms of moment magnitude, MW. However, MW is still not commonly calculated for small local events by many national networks. The preferred magnitude type calculated for local earthquakes by Australia’s National Earthquake Alerts Centre is local magnitude, ML. For use in seismic hazard forecasts, magnitude conversion equations are often applied to convert ML to MW. Unless these conversions are time-dependent, they commonly assume that ML estimation has been consistent for the observation period. While Australian-specific local magnitude algorithms were developed from the late 1980s and early 1990s, regional, state and university networks did not universally adopt these algorithms, with some authorities continuing to use Californian magnitude algorithms. Californian algorithms are now well-known to overestimate earthquake magnitudes for Australia. Consequently, the national catalogue contains a melange of contributing authorities with their own methods of magnitude estimation. The challenge for the 2018 National Seismic Hazard Assessment of Australia was to develop a catalog of earthquakes with consistent local magnitudes, which could then be converted to MW. A method was developed that corrects magnitudes using the difference between the original (inappropriate) magnitude formula and the Australian-specific corrections at a distance determined by the nearest recording station likely to have recorded the earthquake. These corrections have roughly halved the rates of ML 4.5 earthquakes in the Australian catalogue. To address ongoing challenges for catalog improvement, Geoscience Australia is digitising printed and hand-written observations preserved on earthquake data sheets. Once complete, this information will provide a valuable resource that will allow for further interrogation of pre-digital data and enable refinement of historical catalogs. Presented at the 2019 Seismological Society of America Conference, Seattle in the special session on “Seismology BC(d)E: Seismology Before the Current (digital) Era”

Since the publication of the Global Seismic Hazard Assessment Project (GSHAP) hazard map in 1999, Australia has stood out as a region of high earthquake hazard among its stable continental region (SCR) peers. The hazard map underpinning the GSHAP traces its lineage back to the 1990 assessment of Gaull and others. This map was modified through a process of expert judgement in response to significant Australian earthquakes (notably the MW 6.2, 6.3 and 6.6 1988 Tennant Creek sequence and the deadly 1989 MW 5.4 Newcastle earthquake). The modified map, developed in 1991 (McCue and others, 1993), underpins Standards Australia’s structural design actions to this day (AS1170.4–2007). But does this assessment make sense with our current understanding of earthquake processes in SCRs? Geoscience Australia (GA) have embarked to update the seismic hazard model for Australia through the National Seismic Hazard Assessment (NSHA18) project. Members of the Australian seismological community were solicited to contribute alternative seismic source models for consideration as inputs to the updated Australian NSHA18. This process not only allowed for the consideration of epistemic uncertainty in the hazard model in a more comprehensive and transparent manner, but also provides the community as a whole ownership of the final model. The 3rd party source models were assessed through an expert elicitation process that weighed the opinion of each expert based on their knowledge and ability to judge relevant uncertainties. In total, 19 independent seismic source models (including regional and background area sources, smoothed seismicity and seismotectonic sources) were considered in the complete source model. To ensure a scientifically rigorous, transparent and quality product, GA also established a Scientific Advisory Panel to provide valuable and ongoing feedback during the development of the NSHA18. The NSHA18 update yields many important advances on its predecessors, including: calculation in a full probabilistic framework using the OpenQuake-engine; consistent expression of earthquake magnitudes in terms of MW; inclusion of epistemic uncertainty through the use of third-party source models; inclusion of a national fault-source model based on the Australian Neotectonic Features database; inclusion of epistemic uncertainty on fault occurrence models and earthquake clustering; and the use of modern ground-motion models. The preliminary NSHA18 design values are significantly lower than those in the current (1991-era) AS1170.4–2007 map at the 10% in 50-year probability level. However, draft values at lower probabilities (i.e., 2% in 50-years) are entirely consistent (in terms of the percentage land mass exceeding different PGA thresholds) with other SCRs with low strain rates (e.g. the central & eastern United States). The large reduction in seismic hazard at the 10% in 50-year probability level has led to much consternation amongst the building code committee in terms of whether the new draft design values will allow enough resilience to seismic loads. This process underscores the challenges in developing national-scale PSHAs in slowly deforming regions, where 10% in 50-year probability level may not adequately capture the maximum considered earthquake ground motions. Consequently, a robust discussion is required is amongst the Australian building code committee (including hazard practitioners) to determine alternative hazard and/or risk objectives that could be considered for future standards. Presented at the Probabilistic Seismic Hazard Assessment (PSHA) Workshop 2017, Lenzburg, Switzerland

We present earthquake ground motions based upon a paleoseismically-validated characteristic earthquake scenario for the ~ 48 km-long Avonmore scarp, which overlies the Meadow Valley Fault, east of Bendigo, Victoria. The results from the moment magnitude MW 7.1 scenario earthquake indicate that ground motions are sufficient to be of concern to nearby mining and water infrastructure. Specifically, the estimated median peak ground acceleration (PGA) exceeds 0.5 g to more than ~ 10 km from the source fault, and a 0.09 g PGA liquefaction threshold is exceeded out to approximately 50-70 kilometres. Liquefaction of susceptible materials, such as mine tailings, may occur to much greater distances. Our study underscores the importance of identifying and characterising potentially active faults in proximity to high failure-consequence dams, including mine tailings dams, particularly in light of the requirement to manage tailing dams for a prolonged period after mine closure. Paper presented at Australian National Committee on Large Dams (ANCOLD) conference 2020, online. (https://leishman.eventsair.com/ancold-2020-online/)

Geoscience Australia, together with contributors from the wider Australian seismology community, have produced a draft National Seismic Hazard Assessment (NSHA18), recommended for inclusion in the 2018 update of Standards Australia’s Structural design actions, part 4: Earthquake actions in Australia, AS1170.4–2007 (Standards Australia, 2007). This Standard is prepared by Subcommittee BD-006-11, General Design Requirements and Loading on Structures of Standards Australia. The provisional seismic hazard values presented in this report have been submitted to comply with Standards Australia’s public comment and publication timelines. This report provides a brief overview of provisional mean peak ground acceleration values (equivalent to the seismic hazard factor Z in AS1170.4) and the approaches used. The hazard values are calculated on rock sites (AS1170.4 Site Class Be) for a probability of exceedance of 10% in 50 years (0.0021 per annum). Continued refinement of these values will occur throughout, and in response to, the first public comment period. While only minor changes are expected, the final NSHA18 will be completed prior to Standard Australia’s planned second public comment period (likely in late 2017). The NSHA18 update yields many important advances on its predecessors, including: • calculation in a full probabilistic framework (e.g., Cornell, 1968) using the Global Earthquake Model Foundation’s OpenQuake-engine (Pagani et al., 2014); • consistent expression of earthquake magnitudes in terms of moment magnitude, MW; • inclusion of epistemic uncertainty through the use of third-party source models contributed by the Australian seismology community; • inclusion of epistemic uncertainty on magnitude-frequency distributions; • inclusion of a national fault-source model based on the Australian Neotectonic Features database (Clark et al., 2012; Clark et al., 2016); • inclusion of epistemic uncertainty on fault-slip-model magnitude-frequency distributions and earthquake clustering; and • use of modern ground-motion models.

<div>COMET (The Centre for Observation and Modelling of Earthquakes, Volcanoes and Tectonics) uses satellite measurements alongside ground-based observations and geophysical models to study active faults and earthquakes. This talk provides an overview of COMET research products in Türkiye and Central Asia, where interseismic deformation and active faults are directly observable. It also touches on how these products highlight the complexity and difficulty of seismic hazard modelling in Australia.&nbsp;</div><div>Three COMET datasets will be discussed, which each contribute to seismic hazard models. Researchers at COMET have and continue to pioneer INSAR methods including co-seismic interferograms and time-series modelling. For example, the Türkiye (Türkiye) INSAR strain-rate map directly estimates strain-accumulation across faults, while the LICSAR portal and satellite cross-correlation methods are used to quantify co-seismic and post-seismic deformation (including after the devastating 2023 Türkiye-Syria earthquake).&nbsp;</div><div>Similar methods are applied in the Tien Shan, where active faults are identifiable in satellite imagery and elevation data, but rates of activity are uncertain and expensive to obtain through field work. Here COMET and GEM (the Global Earthquake Model) are collaborating to produce block-model informed PSHA inputs using active fault databases, GNSS, and INSAR.&nbsp;</div><div>While these methods are useful in tectonically active regions, they serve to highlight the difficulties facing Australian seismic hazard modelling where similar methods cannot be used due to low (to unobservable) tectonic strain and very long fault recurrence.&nbsp;</div> This paper was presented to the 2023 Australian Earthquake Engineering Conference 23-25 November 2023 (https://aees.org.au/aees-conference-2023/)

<div>The city of Lae is Papua New Guinea (PNG)’s second largest, and is the home of PNG’s largest port. Here, a convergence rate of ~50 mm/yr between the South Bismarck Plate and the Australian Plate is accommodated across the Ramu-Markham Fault Zone (RMFZ). The active structures of the RMFZ are relatively closely spaced to the west of Lae. However, the fault zone bifurcates immediately west of the Lae urban area, with one strand continuing to the east, and a second strand trending southeast through Lae City and connecting to the Markham Trench within the Huon Gulf. </div><div>The geomorphology of the Lae region relates to the interaction between riverine (and limited marine) deposition and erosion, and range-building over low-angle thrust faults of the RMFZ. Flights of river terraces imply repeated tectonic uplift events; dating of these terraces will constrain the timing of past earthquakes and associated recurrence intervals. Terrace riser heights are typically on the order of 3 m, indicating causative earthquake events of greater than magnitude 7. </div><div>Future work will expose the most recently active fault traces in trenches to assess single event displacements, and extend the study to the RMFZ north of Nadzab Airport. These results will inform a seismic hazard and risk assessment for Lae city and surrounding region.</div> Presented at the 2023 Australian Earthquake Engineering Society (AEES) Conference