Ameri, Gabriele

Seismic structural risk analysis of critical facilities may require nonlinear dynamic analysis for which record selection is one of the key issues. Notwithstanding the increasing availability of database of strong-motion records, it may be hard to find accelerograms that fit a specific scenario (e.g., in terms of magnitude and distance) resulting from hazard assessment at the site of interest. A possible, alternative, approach can be the use of artificial and/or simulated ground motion in lieu of real records. Their employment requires systematic engineering validation in terms of structural response and/or seismic risk. Prediction equations for peak and cyclic inelastic single degree of freedom systems’ response, based on Italian accelerometric data, are discussed in this study as a possible benchmark, alongside real record counterparts, for the validation of synthetic records. Even if multiple events would be in principle required, an extremely preliminary validation is carried out considering only four simulated records of the 1980 Irpinia (southern Italy) Mw 6.9 earthquake. Simulated records are obtained through a broadband hybrid integral-composite technique. Results show how this simulation method may lead to generally acceptable results. It is also emphasized how this kind of validation may provide additional results with respect to classical signal-to-signal comparison of real and simulated records.

This article presents a set of Ground-Motion Prediction Equations (GMPEs) for Europe and the Middle East, derived from the RESORCE strong motion data bank, following a standard regression approach. The parametric GMPEs are derived for the peak ground acceleration, peak ground velocity, and 5 %-damped pseudo-absolute acceleration response spectra computed over 23 periods between 0.02 and 3 s, considering the average horizontal-component ground-motions. The GMPEs are valid for distances less than 300 km, hypocentral depth up to 35 km and over the magnitude range 4–7.6. Two metrics for the source-to-station distance (i.e. Joyner-Boore and hypocentral) are considered. The selected dataset is composed by 2,126 recordings (at a period of 0.1 s) related to 365 earthquakes, that includes strong-motion data from 697 stations.The EC8 soil classification (four classes from A to D) discriminates recording sites and four classes (normal, reverse, strike-slip, and unspecified) describe the style of faulting. A subset which contains only stations with measured Vs30 and earthquakes with specified focal mechanism (1,224 records from 345 stations and 255 earthquakes) is used to test of the accuracy of the median prediction and the variability associated to the broader data set. A random effect regression scheme is applied and bootstrap analyses are performed to estimate the 95 % confidence levels for the parameters. The total standard deviation sigma is decomposed into between-events and within-event components, and the site-to-site component is evaluated as well. The results show that the largest contribution to the total sigma is coming from the within-event component. When analyzing the residual distributions, no significant trends are observed that can be ascribed to the earthquake type (mainshock-aftershock classification) or to the non-linear site effects. The proposed GMPEs have lower median values than global models at short periods and large distances, while are consistent with global models at long periods (T>1) s. Consistency is found with two regional models developed for Turkey and Italy, as the considered dataset is dominated by waveforms recorded in these regions.

This article presents comparisons among the five ground-motion models described in other articles within this special issue, in terms of data selection criteria, characteristics of the models and predicted peak ground and response spectral accelerations. Comparisons are also made with predictions from the Next Generation Attenuation (NGA) models to which the models presented here have similarities (e.g. a common master database has been used) but also differences (e.g. some models in this issue are nonparametric). As a result of the differing data selection criteria and derivation techniques the predicted median ground motions show considerable differences (up to a factor of two for certain scenarios), particularly for magnitudes and distances close to or beyond the range of the available observations. The predicted influence of style-of-faulting shows much variation among models whereas site amplification factors are more similar, with peak amplification at around 1s. These differences are greater than those among predictions from the NGA models. The models for aleatory variability (sigma), however, are similar and suggest that ground-motion variability from this region is slightly higher than that predicted by the NGA models, based primarily on data from California and Taiwan.

ShakeMap package uses empirical Ground Motion Prediction Equations (GMPEs) to estimate the ground motion where recorded data are not available. Recorded and estimated values are then interpolated in order to produce a shaking map associated with the seismic event of interest. The ShakeMap approach better works in regions with dense stations coverage, where the observed ground motions adequately constrain the interpolation. In poorly instrumented regions, the ground motion estimate mainly relies on the GMPE, that account only for average characteristics of source and wave propagation processes. In this study we investigated the improvement of ShakeMap in the near fault area when including synthetic estimates. We focus on the 2008, Mw 7.0, Iwate-Miyagi Nairiku (Japan) earthquake as a case study because recorded by a huge number of stations. As first we calculated the shakemaps to be used as reference maps and then removed several subsets of stations from the original data-set, replacing them with: (i) the estimations of the ground motion obtained by using a specific GMPE valid for that area, using simple source information such as the earthquake magnitude and fault geometry; (ii) the peak values from synthetic time-histories computed with a hybrid deterministic-stochastic method for extended fault, using the rupture fault model obtained from the kinematic source inversion of strong-motion records. We evaluate the deviations from the reference map and the sensitivity to the number of sites where recordings are not available. Our results show that shakemaps are more and more reliable as the coverage of stations is dense and uniformly distributed in the near-source area. Moreover, the synthetics account for propagation and source properties in a more correct way than GMPE, and largely improve the results. The hybrid maps reach good fitting levels especially when synthetics are used to integrate real data and for particular strong-motion parameters and stations’ distribution.

On 20 May 2012, at 02:03:52 GMT, an earthquake with Mw 6.1 (RCMT, http://www.bo.ingv.it/RCMT) occurred in northern Italy striking a densely populated area. The mainshock was followed a few hours later by two severe aftershocks having the same local magnitude (Ml 5.1, 1 and 2 in Figure 1a), and by hundreds of smaller aftershocks. Nine days later, on 29 May, at 07:00:03 GMT, a second event with moment magnitude Mw 6.0 (RCMT, http://www.bo.ingv.it/RCMT) occurred to the west, on an adjacent fault segment. This event was also followed by hundreds of aftershocks, three of them having local magnitude 5.3, 5.2 and 5.1 (3, 4 and 5, respectively, in Figure 1a) (locations from Istituto Nazionale di Geofisica e Vulcanologia, hereinafter INGV, http://iside.rm.ingv.it/; Malagnini et al., 2012; Scognamiglio et al., 2012). Despite the moderate number of casualties if compared to other major events in the Italian history, the economic loss was extremely high, resulting in about EUR 5 billion (AON Benfield, 2012, http://www.aon.com/), as the majority of Italian industrial activities and infrastructures concentrate in this area, the eastern Po plain, which is the largest sedimentary basin in Italy. The mainshocks are associated to two thrust faults with an approximate E-W trend dipping to the South (Figure 1b). The majority of the faults in this region are located in the upper crust, at depths lower than 10 km. The two main shocks are among the strongest earthquakes generated by thrust faults ever recorded in Italy in the instrumental era. The Emilia sequence has been extensively recorded by several strong-motion networks, operating in the Italian territory and neighbouring countries. Some of the networks acquire continuous data streams at their national data centres, which are nodes of EIDA (European Integrated Data Archive, hhtp://eida.rm.ingv.it), a federation of several archives, so that the waveforms can be obtained immediately after the occurrence of an event. Other networks, such as the Italian accelerometric network (RAN), managed by the Italian Department of the Civil Protection (hereinafter DPC), distribute the acceleration waveforms through their web site (http://protezionecivile.gov.it). The data set explored in this study is relative to the six events of the sequence having Ml > 5 (Table 1) and consists in 365 accelerograms recorded within a distance of 200 km from the epicentres, that were provided by the permanent and temporary seismic networks of INGV, the Swiss Seismological Service (SED, http://www.seismo.ethz.ch/index) and the DPC.

The Emilia earthquakes of May 20, 2012 (ML 5.9, INGV; MW 6.11, http://www.bo.ingv.it/RCMT/) and May 29, 2012 (ML 5.8, INGV; MW 5.96, http://www.bo.ingv.it/RCMT/) struck an area that in the national reference seismic hazard model [MPS04; http://zonesismiche.mi.ingv.it, and Stucchi et al. 2011] is characterized by expected horizontal peak ground acceleration (PGA) with a 10% probability of exceedance in 50 years that ranges between 0.10 g and 0.15 g (Figure 1), which is a medium level of seismic hazard in Italy. The strong impact of the earthquakes on a region that is not included among the most hazardous areas of Italy, and the ground motion data recorded by accelerometric networks, have given the impression to the population and the media that the current seismic hazard map is not correct, and thus needs to be updated. Since the MPS04 seismic hazard model was adopted by the current Italian building code [Norme Tecniche per le Costruzioni 2008, hereafter termed NTC08; http://www.cslp. it/cslp/] as the basis to define seismic action (the design spectra), any modification to the seismic hazard model would also affect the building code. The aim of this paper is to briefly present the data that support the seismic hazard model in the area, and to perform some comparisons between recorded ground motion with seismic hazard estimates and design spectra. All of the comparisons presented in this study are for the horizontal components only, as the Italian hazard model did not perform any estimates for the vertical component.

In Italy, strong-motion monitoring was started in 1972 by different Institutions, although mainly through Ente Nazionale per l'Energia Elettrica (ENEL; Italian National Electricity Company) and Dipartimento della Protezione Civile (DPC; Italian Department of Civil Protection), with different purposes. These included permanent acceleromet- ric monitoring and temporary monitoring during seismic se- quences or before permanent installation. Today, the National Accelerometric Network (RAN; Rete Accelero- metrica Nazionale) [Gorini et al. 2010, Zambonelli et al. 2011] is operated by the DPC and consists of 464 digital sta- tions. These are distributed throughout the whole national territory, with a prevalence for areas of major seismicity. In 2006, the INGV began strong-motion monitoring, by installing 22 accelerometric stations in northern Italy (RAIS; Rete Accelerometrica Italia Settentrionale; Accelerometric Network of Northern Italy; http://rais.mi.ingv.it/). In 2008, the monitoring was extended to a national scale: this effort led to the installation of 105 accelerometers, collocated with the velocimetric sensors, in selected Rete Sismica Nazionale (RSN; National Seismic Network) sites [Amato and Mele 2008] that are managed by the Centro Nazionale Terremoti (CNT; National Earthquake Centre). Overall, the 127 strong- motion stations that form the INGV Italian strong-motion network homogeneously cover the whole Italian territory. The progress achieved in Italy in the field of strong-mo- tion monitoring and strong-motion data archiving and dis- semination was illustrated in a recently published special issue of the Bulletin of Earthquake Engineering [Luzi et al. 2010]. The strong-motion data recorded by the RAN have been distributed and are available on request to the DPC and to the Italian Accelerometric Archive (ITACA), as the Italian strong-motion database (http://itaca.mi.ingv.it/) [Pacor et al. 2011a], which has been updated with records to 2009. The INGV strong-motion data are archived in real-time and dis- tributed through the European Integrated Data Archive (EIDA; http://eida.rm.ingv.it/) web portal. Recently, an INGV working group developed the first version of a web portal with the aim of archiving, processing and distributing accelerometric data recorded by permanent and temporary INGV stations. This web portal (www.mi. ingv.it/ISMD/; Figure 1, top panel) is composed of two main modules: the former is known as the INGV Strong Motion Data (ISMD, www.mi.ingv.it/ISMD/ismd.h tml/; Figure 1, bottom left panel) and has as its main scope the analyse and distribution in quasi-real time (a few hours after event oc- currence) of the uncorrected accelerometric data, and the related metadata obtained after an automatic processing pro- cedure. This latter, known as the Dynamic Archive (DYNA, http://dyna.mi.ingv.it/DYNA-archive/; Figure 1, bottom right panel) is a dynamic database where manually post- processed accelerometric waveforms are provided, together with their metadata. Both of these archives are designed and structured in such a way that their compilations and updat- ing will be almost completely automatic. At the end of May 2012, a first prototype of the ISMD module was published, providing the uncorrected strong- motion data recorded by the INGV stations for the main events of the Emilia seismic sequence [Massa et al. 2012].

L’INGV, a partire dal 2006, ha iniziato una fase di potenziamento del monitoraggio accelerometrico, installando nelle aree centrali della pianura padana 22 sensori strong-motion (Rete Accelerometrica Italia Settentrionale, RAIS, http://rais.mi.ingv.it/). Dal 2008, sensori accelerometrici sono stati via via installati in 105 siti a Rete Sismica Nazionale (RSN), gestita dal Centro Nazionale Terremoti (CNT). Nel complesso le 127 stazioni accelerometriche presenti sul territorio nazionale costituiscono a tutti gli effetti la rete accelerometrica nazionale INGV. I dati acquisiti da tutte le stazioni accelerometriche sono attualmente distribuiti in tempo reale tramite il portale EIDA (European Integrated Data Archive; http://eida.rm.ingv.it/) e sono principalmente utilizzati per il calcolo delle Shakemaps a scala nazionale. Attualmente, l’INGV sta realizzando un portale web per la distribuzione dei dati accelerometrici registrati dalle stazioni INGV, composto da 2 moduli distinti: il primo, denominato ISMD, ha lo scopo di archiviaziare e distribuire in tempo quasi reale (poche ore dopo l’evento) le forme d’onda accelerometriche in formato non corretto ed i relativi metadati ottenuti a seguito di una procedura di processamento automatico; il secondo, denominato DYNA, è una banca dati relazionale, contenente le forme d’onda di accelerazione, velocità e spostamento e gli spettri di risposta di accelerazione, ottenuti attraverso il processamento manuale dei segnali non corretti, oltre ai relativi metadati associati agli eventi sismici ed alle stazioni di registrazione Il prototipo del portale dei dati accelerometrici INGV (Figura 1) è stato pubblicato lo scorso maggio, a seguito della sequenza sismica Emiliana.

Near-fault strong-ground motions (0.1–10 Hz) recorded during the Mw 6.3 2009 L’Aquila earthquake exhibit great spatial variability. Modeling the observed seismograms allows linking distinct features of the observed wavefield to particular source and propagation effects and provides insights on strong motion complexity from this moderate magnitude event. We utilize a hybrid integral-composite approach based on a k-square kinematic rupture model, combining low-frequency coherent and high-frequency incoherent source radiation and providing omega-squared source spectral decay. Several source model features, proven to be stable by means of an uncertainty analysis in the preceding low-frequency (<0.2 Hz) multiple finite-extent source inversion (Paper 1), were constrained. Synthetic Green’s functions are calculated in a 1D-layered crustal model including 1D soil profiles to account for site-specific response (where available). The results show that although the local site effects improve the modeling, the spatial broadband ground-motion variability is to large extent controlled by the rupture kinematics. The modeling thus confirms and further constraints the source model features, including the position and slip amount of the two main asperities, the largest asperity time delay and the rupture velocity distribution on the fault. Furthermore, we demonstrate that the crossover frequency dividing the coherent and incoherent wavefield, often considered independent on the station position, has to be variable in order to adequately reproduce both near and far station recordings. This suggests that the incoherency of the radiated wavefield is controlled by the wave-propagation phenomena and/or the initial updip rupture propagation was very smooth (coherent) up to relatively high frequencies (>2 Hz)

In Italy in the last years many ground motion prediction equations (hereinafter GMPEs) were calibrated both at national and regional scale using weak and strong motion data recorded in the last 30 years by several networks. Moreover many of the Italian strongest earthquakes were included in global datasets in order to calibrate GMPEs suitable to predict ground-motion at very large scale. In the last decade the Sabetta and Pugliese (1996) relationships represented a reference for the ground motion predictions in Italy. At present all Italian strong-motion data, recorded from 1972 by RAN (Italian Accelerometric Network), and more recently by other regional networks (e.g. RAIS, Strong motion network of Northern Italy), are collected in ITACA (ITalian ACcelerometric Archive). Considering Italian strong-motion data with Mw  4.0 and distance (Joyner-Boore or epicentral) up to 100 km, new GMPEs were developed by Bindi et al. (2009), aimed at replacing the older Italian relationships. The occurrence of the recent 23rd December 2008, Mw 5.4, Parma (Northern Italy) earthquake and the 6th April 2009, Mw 6.3, L’Aquila earthquake, allowed to upgrade the ITACA data set and gave us the possibility to validate the predictive capability of many GMPEs, developed using Italian, European and global data sets. The results are presented in terms of quality of performance (fit between recorded and predicted values) using the maximum likelihood approach as explained in Spudich et al. (1999). Considering the strong-motion data recorded during the L’Aquila sequence the considered GMPEs, in average, overestimate the observed data, showing a dependence of the residuals with distance in particular at higher frequencies. An improvement of fit is obtained comparing all Italian strong-motion data included in ITACA with the European GMPEs calibrated by Akkar and Bommer (2007 a,b) and the global models calibrated by Cauzzi and Faccioli (2008). In contrast, Italian data seem to attenuate faster than the NGA models calibrated by Boore and Atkinson (2008), in particular at higher frequencies.