Rupture directivity effects in ground motion are known since many years to both seismologists and earthquake engineers, i.e. in sites that are in a particular geometrical configuration with respect to the rupture, the velocity fault-normal signals may show a large pulse which occurs at the beginning of the record and contains the most of energy. The results are waveforms different from ordinary ground motions recorded in the far field or in geometrical conditions not favorable with respect to directivity. Current attenuation laws are not able to capture such effect well, if at all, and current probabilistic seismic hazard analysis is not able to predict the resulting peculiar spectral shape. Moreover, it is believed that structures with dynamic behavior in a range of periods related to the pulse period may be subjected to underestimated seismic demand. In the paper this is investigated and increments in both elastic and inelastic seismic actions are quantified using a large dataset of records, from the next generation attenuation project (NGA), in which a fraction is comprised of velocity pulses identified in other studies. These analyses employ recently developed tools and procedures to assess directivity effects and to quantify the associated threat in terms of seismic action on structures. Subsequently, the same tools are used in one of the first attempts to identify near-source effects in the data recorded during a normal faulting earthquake, the mainshock of the recent Abruzzo (central Italy) sequence, leading to conclude that pulse-like effects are likely to have occurred in the event, that is (1) observation of pulse-like records in some near-source stations is in fair agreement with existing predictive models, (2) the increment in seismic demand shown by pulse-like ground motion components complies with the results of the analysis of the NGA data, and (3) seismic demand in non-impulsive recordings is generally similar to what expected for ordinary records. The results may be useful as a benchmark for inclusion of near-source effect in design values of seismic action and structural risk analysis. || Note that this modification of attenuation may be considered as a narrow band one [13], i.e. elastic demand is magnified for selected frequency corresponding to the pulse period. Former models to modify ordinary GMPEs to account for directivity are broad band; e.g. that of Somerville et al. [1]. Statistical tests on integral parameters distributions would be useful to confirm or not these comparisons, but the large sample sizes and asymmetric shape distributions (see Figure 4) suggest that the most parametric tests would not be useful, while non-parametric are not as powerful.Results of seismological studies have shown that the Abruzzo event was a normal faulting earthquake (or dip-slip), with a rectangular rupture plane of about 17×14 km 2 and located at a depth ‡ ‡ ‡ In the application of the attenuation relationship [25], the surface wave magnitude (M s ) value of the L'Aquila earthquake is assumed to be the...

Near-source pulse-like records resulting from rupture's directivity have been found to depart from so-called ordinary ground motions in terms of both elastic and inelastic structural seismic demands. In fact, response spectra may be strong if compared with what is expected from common ground motion prediction equations. Moreover, because not all spectral ordinates are affected uniformly, a peculiar spectral shape, with an especially amplified region depending on the pulse period, may follow. Consequently, inelastic seismic demand may show trends different to records not identified as pulse-like (i.e., ordinary). This latter aspect is addressed in the study reported in this short communication, where a relatively large dataset of identified impulsive near-source records is used to derive an analytical-form relationship for the inelastic displacement ratio. It is found that, similar to what was proposed in literature for soft soil sites, a double-opposite-bumps form is required to match the empirical data as a function of the structural period over the pulse period ratio. The relationship builds consistently on previous studies on the topic, yet displays different shape with respect to the most common equations for static structural assessment procedure

Earthquakes damage engineering structures near, relatively to the rupture's size, to the source. In this region, the fault's dynamics affect ground motion propagation differently from site to site, resulting in systematic spatial variability known as directivity. Although a number of researches recommend that records with directivity-related velocity pulses should be explicitly taken into account when defining design seismic action on structures, probabilistic seismic hazard analysis (PSHA), in its standard version, seems inadequate for the scope. In the study, it is critically reviewed why, from the structural engineering point of view, hazard assessment should account for near-source effects (i.e., pulse-like ground motions), and how this can be carried out adjusting PSHA analytically via introduction of specific terms and empirically calibrated models. Disaggregation analysis and design scenarios for near-source PSHA are also formulated. The analytical procedures are then applied to develop examples of hazard estimates for sites close to strike–slip or dip–slip faults and to address differences with respect to the ordinary case, that is, when pulse-like effects are not explicitly accounted for. Significant increase of hazard for selected spectral ordinates is found in all investigated cases; increments depend on the fault-site configuration. Moreover, to address design scenarios for seismic actions on structures, disaggregation results are also discussed, along with limitations of current design spectra to highlight the pulse-like effects of structural interest. Finally, an attempt to overcome these, by means of disaggregation-based scenarios specific for the pulse occurrence case, is presented

No abstract