Search

This paper provides an overview and introduction to the development of non-ergodic ground-motion models, GMMs. It is intended for a reader who is familiar with the standard approach for developing ergodic GMMs. It starts with a brief summary of the development of ergodic GMMs and then describes different methods that are used in the development of non-ergodic GMMs with an emphasis on Gaussian process (GP) regression, as that is currently the method preferred by most researchers contributing to this special issue. Non-ergodic modeling requires the definition of locations for the source and site characterizing the systematic source and site effects; the non-ergodic domain is divided into cells for describing the systematic path effects. Modeling the cell-specific anelastic attenuation as a GP, and considerations on constraints for extrapolation of the non-ergodic GMMs are also discussed. An updated unifying notation for non-ergodic GMMs is also presented, which has been adopted by the authors of this issue. [ABSTRACT FROM AUTHOR]

A new approach for creating a non-ergodic pseudo-spectral acceleration (PSA) ground-motion model (GMM) is presented, which accounts for the magnitude dependence of the non-ergodic effects. In this approach, the average PSA scaling is controlled by an ergodic PSA GMM, and the non-ergodic effects are captured with non-ergodic PSA factors, which are the adjustment that needs to be applied to an ergodic PSA GMM to incorporate the non-ergodic effects. The non-ergodic PSA factors are based on the effective amplitude spectrum (EAS) non-ergodic effects and are converted to PSA through Random Vibration Theory (RVT). The advantage of this approach is that it better captures the non-ergodic source, path, and site effects through small-magnitude earthquakes. Due to the linear properties of the Fourier Transform, the EAS non-ergodic effects of the small events can be applied directly to the large magnitude events. This is not the case for PSA, as response spectra are controlled by a range of frequencies, making PSA non-ergodic effects dependent on the spectral shape, which in turn is magnitude-dependent. Two PSA non-ergodic GMMs are derived using the ASK14 (Abrahamson et al. in Earthq Spectra 30:1025–1055, 2014) and CY14 (Chiou and Youngs in Earthq Spectra 30:1117–1153, 2014) GMMs as backbone models, respectively. The non-ergodic EAS effects are estimated with the LAK21 (Lavrentiadis et al. in Bull Earthq Eng) GMM. The RVT calculations are performed with the V75 (Vanmarcke in ASCE Mech Eng Mech Division 98:425–446, 1972) peak factor model, the D a 0.05 - 0.85 estimate of AS96 (Abrahamson and Silva in Apendix A: empirical ground motion models, description and validation of the stochastic ground motion model. Tech. rep.,. Brookhaven National Laboratory, New York) for the ground-motion duration, and BT15 (Boore and Thompson in Bull Seismol Soc Am 105:1029–1041, 2015) oscillator-duration model. The California subset of the NGAWest2 database (Ancheta et al. in Earthq Spectra 30:989–1005, 2014) is used to fit both models. The total aleatory standard deviation of each of the two non-ergodic PSA GMMs is approximately 25 % smaller than the total aleatory standard deviation of the corresponding ergodic PSA GMMs. This reduction has a significant impact on hazard calculations at large return periods. In remote areas, far from stations and past events, the reduction of aleatory variability is accompanied by an increase in epistemic uncertainty. [ABSTRACT FROM AUTHOR]

A new non-ergodic ground-motion model (GMM) for effective amplitude spectral (EAS) values for California is presented in this study. EAS, which is defined in Goulet et al. (Effective amplitude spectrum (eas) as a metric for ground motion modeling using fourier amplitudes, 2018), is a smoothed rotation-independent Fourier amplitude spectrum of the two horizontal components of an acceleration time history. The main motivation for developing a non-ergodic EAS GMM, rather than a spectral acceleration GMM, is that the scaling of EAS does not depend on spectral shape, and therefore, the more frequent small magnitude events can be used in the estimation of the non-ergodic terms. The model is developed using the California subset of the NGAWest2 dataset (Ancheta in PEER NGA-West2 database. Tech. rep., PEER, Berkeley, CA, 2013). The Bayless and Abrahamson (Bull Seismol Soc Am 109(5): 2088-2105, https://doi.org/10.1785/0120190077, 2019b) (BA18) ergodic EAS GMM was used as backbone to constrain the average source, path, and site scaling. The non-ergodic GMM is formulated as a Bayesian hierarchical model: the non-ergodic source and site terms are modeled as spatially varying coefficients following the approach of Landwehr et al. (Bull Seismol Soc Am 106(6):2574-2583. https://doi.org/10.1785/0120160118, 2016), and the non-ergodic path effects are captured by the cell-specific anelastic attenuation attenuation following the approach of Dawood and Rodriguez-Marek (Bull Seismol Soc Am 103(2B):1360-1372, https://doi.org/10.1785/0120120125, 2013). Close to stations and past events, the mean values of the non-ergodic terms deviate from zero to capture the systematic effects and their epistemic uncertainty is small. In areas with sparse data, the epistemic uncertainty of the non-ergodic terms is large, as the systematic effects cannot be determined. The non-ergodic total aleatory standard deviation is approximately 30 to 40 % smaller than the total aleatory standard deviation of BA18. This reduction in the aleatory variability has a significant impact on hazard calculations at large return periods. The epistemic uncertainty of the ground motion predictions is small in areas close to stations and past events. [ABSTRACT FROM AUTHOR]

Catastrophic earthquakes in Uganda have the potential for detrimental consequences on the socio-economic welfare and resilience of communities. Despite considerable efforts in predicting earthquake risk across Africa, a national comprehensive seismic risk study for Uganda does not exist. With increasing population, urbanisation and rapid construction, seismic risk is escalating fast and is compounded by the high vulnerability of buildings and scanty disaster prevention and mitigation strategies. This study uses the probabilistic event-based risk calculator of the OpenQuake-engine to assess potential risks resulting from future earthquakes. Although the building exposure model is largely inferred and projected from the national population and housing census of 2014, total replacement costs are obtained by performing series of interviews with local engineering practitioners. Analytical vulnerability curves are selected from Global Earthquake Model (GEM) database. Seismic hazard studies confirm that western Uganda is exposed to the highest level of seismicity where peak ground accelerations on rock ground can reach up to 0.27 g over a 475-year return period. Relative to Uganda’s gross domestic product, the associated seismic risk estimates indicate mean economic loss ratios of 0.36%, 2.72% and 4.94% over 10, 50 and 100-year return periods respectively; with mean annual economic loss of US$ 74.7 million (0.34% relative to the total replacement value) and annual deaths averaging 71 persons across the whole country. It is envisaged that the findings will inform strategic land use planning patterns, earthquake insurance pricing and foster the continuous improvement of Uganda’s National Policy for Disaster Preparedness and Management. [ABSTRACT FROM AUTHOR]

Probabilistic seismic hazard analyses (PSHA) require that at least the mean activity rate be known, as well as the distribution of magnitudes. Within the Gutenberg–Richter assumption, the magnitudes follow an exponential distribution which is upperly truncated to a maximum possible magnitude denoted m max . This parameter is often fixed from expert judgement under seismo-tectonic considerations, due to a lack of universal method. This paper proposes an alternative to the common Gutenberg Richter model based on the extreme value theory: it models the tail distribution of the magnitudes with a generalized Pareto distribution (GPD). To integrate this GPD model in a PSHA calculation for areas of low to moderate seimsicity, we propose the Randomized Gutenberg–Richter model: this is an innovative approach based on the usual exponential distribution where m max is randomized and follows a distribution defined from that previous GPD model. The inference process takes into account the time varying level of completeness of the catalogue and the uncertainty in the magnitude value itself. In order to quantify the impacts of the new model on the seismic hazard levels, we implemented it into a realistic probabilistic seismic hazard calculation. Results indicate that the Randomized Gutenberg–Richter model seems to be very useful for future PSHA models. This new recurrence model avoids having to fix a priori the maximum possible magnitude m max . This is particularly interesting as there is no widely accepted method to estimate that upper bound. [ABSTRACT FROM AUTHOR]

Seismic hazard maps from probabilistic seismic hazard analysis or PSHA collect, at different sites, the values of the (site-specific) ground motion intensity measures of interest that, taken individually, have the same exceedance return period. For large-scale analyses, a widely used intensity measure is the macroseismic (MS) intensity, that provides an assessment of the earthquake effect based on the observed consequences in the hit area. Hazard maps can be developed in terms of MS intensity, and some examples exist in this respect. In the case of Italy, the last MS hazard map is based on the same seismic source model (known as MPS04) adopted to derive the design seismic actions of the current building code, a study dating more than ten years ago. It provides results in terms of countrywide Mercalli–Cancani–Sieberg (MCS) intensity level with 475 years return period. This short paper presents and discusses MCS probabilistic seismic hazard maps for Italy based on a recent grid-seismicity source model, herein named MPS19, synthetizing the large effort of a wide scientific community. The results, which are obtained by means of classical PSHA, are given in the form of maps referring to the 475 years return period, and also others of earthquake engineering interest. Moreover, it is discussed that the return period does not univocally identifies the MS intensity because, although MS is, by definition, a discrete random variable, it is modelled, in a given earthquake, by means of a normal distribution, that is, treated as continuous. Thus, the maps of the minimum return period causing the occurrence or exceedance of different MCS intensities are also provided. Finally, the comparison between the 475 years return period hazard map presented and the one which is currently the point of reference in Italy, that is, computed using MPS04, is briefly discussed. All the computed maps are made available to the reader as supplemental material. [ABSTRACT FROM AUTHOR]

African seismicity is predominantly localized along the East African Rift System (EARS), which is the major active tectonic feature of Sub-Saharan Africa. Besides the EARS, however, significant seismicity also occurs along a wide belt bounding the Mediterranean coastline. This tectonically active region extends discontinuously from Morocco to Egypt and its activity is controlled by the complex interaction between the Nubian and Eurasian plates, varying from transpression in the Atlas orogen in the west to transtension in the east. A record of large earthquake events is documented for the whole region, some of them causing moderate to severe levels of damage, mostly because of the high vulnerability of local buildings and structures, a condition which is still largely persistent in many areas. Currently, a number of seismic hazard models exist at local and national scales for North Africa, developed within independent projects and created using inhomogeneous data sources and different processing techniques. Unfortunately, such diversity makes their direct comparison problematic, obscuring the differences in seismic hazard across neighbouring areas and preventing the development of comprehensive long-term risk mitigation strategies. In fact, the last effort to produce a homogenized model for the whole Africa continent dates back to the GSHAP project, which is almost 20 years old. The creation of a unique seismic hazard model for North Africa, uniform across countries, is therefore a main concern. Since its inception, the Global Earthquake Model Foundation (GEM) is committed to the creation of a worldwide mosaic of high-quality, reproducible and openly accessible seismic hazard models, uniformly represented using the format adopted by the OpenQuake engine (OQ), a state-of-the-art, free and open-source software package for seismic hazard and risk assessment. We describe the development of a new comprehensive PSHA model for North Africa using GEM tools. The source model combines active faults and distributed seismicity, the former constrained from published geological descriptions and geodetic data, while the latter from the harmonisation of published earthquake catalogues. [ABSTRACT FROM AUTHOR]

The basic goal of earthquake resistant building design codes is to provide life safety in the event of extreme earthquake loading and to reduce the seismic risk in the long run. In the most building codes including Iranian building code (Standard No. 2800), design basis earthquake is defined as an earthquake with 10% probability of exceedance (PE) in 50 years. Such an approach leads to buildings with unknown reliability, meaning that the probability of exceedance from the specified limit states in the service lifetime is not known. Furthermore, the regions with the same mapped peak ground acceleration (PGADBE) and different hazard curves, may have different seismic risks. Therefore, life safety is not uniformly provided across the entire region. Here, an attempt is made to estimate PGADBEs based on the uniform collapse risk approach for Tehran megacity. The building capacity and PGA hazard curve for 354 grids in Tehran are computed. Then, the risk-adjusted design PGA at different grid points are calculated by optimizing risk integral using the acceptance criteria of 1% in 50 years for ordinary usage buildings (risk category II). The computed risk-adjusted design PGAs vary from 0.33 g in the east to about 0.45 g toward west, south, and north. Meanwhile, the proposed mapped PGADBE in standard 2800 is a constant value of 0.35 g for the whole region. To preserve a uniform collapse risk, risk-adjustment factors to modify the mapped PGADBE are introduced. Furthermore, the maximum considered earthquake (MCE) required to evaluate the acceptance criteria is found to be in the range of 1.5% to 2.5% PE in 50 years. The PGAs with 2% PE in 50 years provide nearly uniform acceptable probability of collapse for buildings classified in risk category II. In addition, it is found that the buildings' seismic importance factor (I) also needs to be revised from 1.2 to 1.3 for seismic risk category III, and from 1.4 to 1.6 for seismic risk category IV to preserve the 0.5% and 0.25% collapse risk in 50 years, respectively. The outcomes also revealed that the sensitivity analysis with regard to the uncertainty of buildings' collapse fragility curve can be ignored for buildings in risk categories I and II, while it needs to be performed for buildings in risk categories III, and especially IV. Finally, the estimated PGA hazard curves of the Tehran megacity are approximated by the cubic functions to expedite the future risk assessment studies in Tehran megacity. [ABSTRACT FROM AUTHOR]

Structural reliability methods are employed in this paper for nationwide probabilistic seismic hazard analysis of Iran and in-depth seismic hazard and hazard sensitivity analysis of Downtown Tehran. The companion paper proposes two analysis approaches based on reliability methods and presents the technical framework for such analyses. The first approach was based on Monte Carlo sampling reliability analysis, which is employed here to conduct hazard analysis at a grid comprising 3695 sites throughout Iran. The results are used to produce nationwide hazard maps for the peak ground acceleration, spectral acceleration, and cumulative absolute velocity. The other analysis approach employs the first- and second-order reliability methods and importance sampling and is applied to the seismic hazard analysis of Downtown Tehran. This analysis identifies the seismic sources that are most influential on the seismicity of Tehran. In addition, reliability sensitivity and importance measures are employed to identify the most influential sources of uncertainty on the seismic hazard at Downtown Tehran. For instance, they identify the geometry of, and the rupture location on, adjacent faults, the shear wave velocity, and the intra-event residuals of ground motion prediction models as the most important sources of uncertainty. [ABSTRACT FROM AUTHOR]

This paper presents a new probabilistic framework for seismic hazard assessment and hazard sensitivity analysis. Hazard in this context means the probability of exceeding a measure of ground shaking intensity, such as peak ground acceleration and spectral acceleration. The main components of the proposed framework include structural reliability methods to estimate exceedance probabilities and their sensitivities, and multiple probabilistic models for earthquake occurrence, magnitude, location, and ground motion. This paper presents two analysis approaches. The first approach utilizes the first- and second-order reliability methods and importance sampling. This approach efficiently yields the hazard exceedance probabilities at a single site. The second approach employs the Monte Carlo sampling reliability method and yields the hazard exceedance probabilities at a multitude of sites in a single analysis, which is suited for large-scale seismic zonation. This paper also presents the probabilistic models that are suited for such analyses with an emphasis on characterization of epistemic uncertainties. Finally, novel sensitivity measures are proposed for hazard sensitivity analysis. These measures provide a framework to identify the most important uncertainties and guide the research to reduce these uncertainties over time. The proposed approach is validated and showcased by an illustrative example. The companion paper presents a comprehensive application to hazard analysis of Iran. [ABSTRACT FROM AUTHOR]

The magnitude frequency distributions (MFDs) are one of the most important components of seismic source modeling in probabilistic seismic hazard analysis (PSHA). They describe the annual occurrence rates of ruptures that are expected to occur in the seismic source. Eventually, the occurrence rates (or probabilities) of ruptures affect the exceedance rate (or probability) and the amplitude of the target ground-motion intensity metric, which is the essential product in PSHA. To this connection, proper portrayal of parametric uncertainty in MFDs entails justifiable ground-motion amplitude exceedance distributions. In this paper, we propose two alternative approaches to account for the parametric uncertainties in the modeling of MFDs. Both approaches treat MFD model parameters as random variables and describe their conditional probability distributions conditioned on seismic source activity. Although both approaches are tailored to structure a proper MFD logic-tree, they differ in the way they handle the conditional probabilities to delineate the parametric uncertainty associated with each MFD model parameter. We first explain the theoretical background of the alternative methods, and then discuss their similarities (and differences) from a case study. [ABSTRACT FROM AUTHOR]

A new seismic hazard model for Cairo, the capital city of Egypt is developed herein based on comprehensive consideration of uncertainties in various components of the probabilistic seismic hazard analysis. The proposed seismic hazard model is developed from an updated catalogue of historical and instrumental seismicity, geodetic strain rates derived from GPS-based velocity-field of the crust, and the geologic slip rates of active faults. The seismic source model consists of area sources and active faults characterised to forecast the seismic productivity in the region. Ground motion prediction models are selected to describe the expected ground motion at the sites of interest. The model accounts for inherent epistemic uncertainties of statistical earthquake recurrence; maximum magnitude; ground motion prediction models, and their propagation toward the obtained results. The proposed model is applied to a site-specific hazard analysis for Kottamiya, Rehab City and Zahraa-Madinat-Nasr (hereinafter referred to as Zahraa) to the East of Cairo (Egypt). The site-specific analysis accounts for the site response, through the parameterization of the sites in terms of average 30-m shear-wave velocity (Vs30). The present seismic hazard model can be considered as a reference model for earthquake risk mitigation and proper resilience planning. [ABSTRACT FROM AUTHOR]

Earthquakes trigger other earthquakes and build up clusters in space and time that in turn create a bias in seismic catalogues. Therefore, declustering is considered as a prerequisite in seismic studies, particularly for probabilistic seismic hazard analysis, not only to eliminate the bias but also to decouple mainshocks and triggered events. However, a declustering process is not a straightforward task due to the complex nature of earthquake phenomena. There exist several declustering methods that mostly employ subjective rules to distinguish between background seismicity and offsprings. Eventually, the final declustered catalogues usually deviate significantly according to the employed method. This issue is raising some concerns, such as how to select the most suitable declustering algorithm, or to assess how this selection affects seismic hazard assessment. In consequence, the main goal of this paper is to quantify the sensitivity of seismic hazard assessments to different declustering techniques. Accordingly, the recently compiled Turkish earthquake catalogue was declustered by making use of three declustering algorithms. A total of six declustered catalogues, two catalogues per method, one by implementing the default input parameters, and one by altering the free input parameters of the employed methods, were produced. The clusters of selected earthquakes were studied in terms of the spatial-temporal distribution of earthquake sequences. A sensitivity analysis was conducted through the major steps of seismic hazard assessment for Istanbul metropolitan city. The seismicity of Istanbul and surroundings was modeled on the basis of four areal source zones. Comparative studies showed that, while the selected declustering algorithm did not significantly affect the completeness periods of moderate to large size earthquakes, it considerably altered those of small magnitude events (e.g. Mw 4.3-5.2) and consequently the recurrence parameters of the source zones. Depending on the declustering algorithm and input parameters, the activity rate was observed to vary up to a factor of two. The differences in the declustered catalogues obtained from different declustering approaches resulted in considerable variations in seismic hazard estimations. The hazard maps at return periods of 475 and 2475 years indicated that peak ground acceleration values may vary up to 20% at some locations. Moreover, the differences in 5% damped elastic spectral accelerations at T = 0.2 for the return periods of 475 and 2475 years are about 18 and 12%, respectively, on the southern shores of Istanbul where the highest hazard levels are observed. [ABSTRACT FROM AUTHOR]

Paper describes the extensive work done in the SIGMA project, aimed at improving knowledge on data, methods and tools to better quantify uncertainties in seismic hazard assessment (SHA). The authors cooperated in the study of potential faults and geological structures, earthquake catalogues, selection of ground motion prediction equations, and methods for site effect evaluation suitable for SHA. All the contributions merged into a probabilistic seismic hazard study conducted for three representative sites of the Po Plain in Northern Italy. Po Plain is a low-to-moderate seismicity region, characterized by some critical features, such as blind faulting and deep alluvium sediments, and by scarcity of strong motion data; these sources of uncertainties in seismic hazard estimation are common to other low seismicity areas around the world. Within SIGMA, special care was devoted to: (a) the use of the single station sigma approach inside the probabilistic SHA, (b) the comparative use of generalized attenuation functions to evaluate the hazard contribution of composite fault systems, and (c) the study of the epistemic uncertainties at play when different modelling approaches to site effects are used. [ABSTRACT FROM AUTHOR]

The East African Rift System is the major active tectonic feature of the Sub-Saharan Africa region. Although the seismicity level of this divergent plate boundary can be described as moderate, several damaging earthquakes have been reported in historical times, and the seismic risk is exacerbated by the high vulnerability of the local buildings and structures. Formulation and enforcement of national seismic codes is therefore an essential future risk mitigation strategy. Nonetheless, a reliable risk assessment cannot be done without the calibration of an updated seismic hazard model for the region. A major limitation affecting the assessment of seismic hazard in Sub-Saharan Africa is the lack of basic information needed to construct source and ground motion models. The historical earthquake record is sparse, with significant variation in completeness over time across different regions. The instrumental catalogue is complete down to sufficient magnitude only for a relatively short time span. In addition, mapping of seismogenically active faults is still an on-going task, and few faults in the region are sufficiently constrained as to allow them to be directly represented within the seismic hazard model. Recent studies have identified major seismogenic lineaments, but there is substantial lack of kinematic information for intermediate-to-small scale tectonic features, information that is essential for the proper calibration of earthquake recurrence models. In this study, we use new data and Global Earthquake Model (GEM) computational tools such as the Hazard Modeller's Toolkit and the OpenQuake engine to perform a pilot study of the seismic hazard associated with the East African Rift. The hazard model obtained has been created using the most recent information available from scientific literature, global bulletins and local earthquake catalogues, including those from AfricaArray projects. In this report, in accordance with the GEM philosophy, we describe in detail all working assumptions, main processing steps, data analyses and interpretations used for the model setup. [ABSTRACT FROM AUTHOR]

The companion paper by Rota and Rosti (Bull Earthq Eng, 2017) illustrates a methodology for comparing PSH results with historical macroseismic observations at different scales, in terms of mean damage. This paper presents examples of application of the methodology, at the different considered scales, to the South-East quarter of France. This moderate seismicity region is characterised by a long history of civilization, which makes macroseismic observations available for a long time span, although they are highly approximated measures of the seismic action and they are affected by significant uncertainties. The first scale of application presented is at a single site, i.e. the city of Annecy. This example shows that, despite the seismic history of this city is characterised by a significant number of macroseismic observations spanning over a long time period, they mostly consist of very low intensity values and hence the comparison at a single site is not very meaningful. Therefore, the comparison is carried out on a set of seven aggregated sites, well distributed in the region of interest, providing interesting results and suggesting the opportunity of extending the comparison at an even larger scale. The comparison at the regional scale also allows some interesting observations, although it is obviously able to only provide general (average) indications on the PSH results. These comparisons were aimed at showing the applicability of the proposed methodology, pointing out advantages and drawbacks of the different application scales. [ABSTRACT FROM AUTHOR]

In this study, we apply an empirical scoring method to evaluate the feasibility of probabilistic seismic hazard analyses at regional scale in Italy accounting for site amplification, which is taken into account through the application of a set of ground motion prediction equations (GMPEs) defined for specific ground types. Precisely, this method calculates the agreement (in terms of likelihood) between the hazard results computed using a specific hazard model and the number of ground-motion exceedances at a set of reference sites. Such a procedure is applied to quantify the likelihood of the outcomes of different hazard models, each based on a specific GMPE, with respect to the observations at 56 accelerometric sites operating in Italy for at least 25 years. Indirectly, this allows evaluating the influence of the selected GMPE in providing reliable hazard estimates. Seven possible GMPEs, applicable in active shallow crustal regions like Italy, have been examined taking into account the correlation among the hazard estimates provided by each computational model at the reference sites. Our results indicate that, although using GMPEs for pre-defined soil categories provides only a broad assessment of the hazard (since it ignores specific site response), large-scale hazard maps of Italy that are compatible with observations can be provided when suitable GMPEs are considered. A restricted number of GMPEs was found appropriate to this scope. [ABSTRACT FROM AUTHOR]

The article presents a comparison of different probabilistic methods for ground motion hazard assessments that include site effects. The approaches examined here were selected and refined during the different phases of the S2-Project, which this journal volume is addressed to. Different procedures characterized by different levels of sophistication, from the simpler one based on the use of standard ground motion predictive equations for specific ground types to the more complex one based on the convolution of a site-specific amplification function (and its variability) with the hazard curve for reference rock, are compared and contrasted with the aim of pointing out strengths and weaknesses of each of them. In addition, a fully non-ergodic approach that separates the epistemic contribution (i.e., the epistemic uncertainty affecting the soil properties) from the total variability in site amplification is presented. To fulfill the scope of the work, the study focuses on three test sites in Italy characterized by different geological conditions and seismicity levels: Mirandola and Soncino in the Po Plain (northern Italy) and Peglio in central Italy. [ABSTRACT FROM AUTHOR]

Ground motion prediction equations (GMPEs) have a major impact on seismic hazard estimates, because they control the predicted amplitudes of ground shaking. The prediction of ground-motion amplitudes due to mega-thrust earthquakes in subduction zones has been hampered by a paucity of empirical ground-motion data for the very large magnitudes (moment magnitude ( M) $$>$$ 7) of most interest to hazard analysis. Recent data from Tohoku M9.0 2011 earthquake are important in this regard, as this is the largest well-recorded subduction event, and the only such event with sufficient data to enable a clear separation of the overall source, path and site effects. In this study, we use strong-ground-motion records from the M9 Tohoku event to derive an event-specific GMPE. We then extend this M9 GMPE to represent the shaking from other M $$>$$ 7 interface events in Japan by adjusting the source term. We focus on events in Japan to reduce ambiguity that results when combining data in different regions having different source, path and site effect attributes. Source levels (adjustment factors) for other Japanese events are determined as the average residuals of ground-motions with respect to the Tohoku GMPE, keeping all other coefficients fixed. The mean residuals (source terms) scale most steeply with magnitude at the lower frequencies; this is in accord with expectations based on overall source-scaling concepts. Interpolating source terms over the magnitude range of 7.0-9.0, we produce a GMPE for large interface events of M7- M9, for NEHRP B/C boundary site conditions (time-averaged shear-wave velocity of 760 m/s over the top 30 m) in both fore-arc and back-arc regions of Japan. We show how these equations may be adjusted to account for the deeper soil profiles (for the same value of $$\hbox {V}_\mathrm{S30})$$ in western North America. The proposed GMPE predicts lower motions at very long periods, higher motions at short periods, and similar motions at intermediate periods, relative to the simulation-based GMPE model of Atkinson and Macias () for the Cascadia subduction zone. [ABSTRACT FROM AUTHOR]