Search

The Community Stress Drop Validation Study has been organized as a technical activity group (TAG) of SCEC (Southern California Earthquake Center) with the aim of investigating the source parameters of the 2019 Ridgecrest seismic sequence in California. Information about the stress drop TAG are available trough the benchmark web-page (https://www.scec.org/research/stress-drop-validation). Several groups applied different techniques to a shared data set with the objective of extracting source parameters (e.g. seismic moment and corner frequency) and in turn to estimate the stress drop. We applied a spectral decomposition approach known as generalized inversion technique (GIT) and the overall analyses are presented in a series of two articles (Bindi et al 2023a; Bindi et al 2023b). Results in the form of files, figures, and tables are disseminated through this archive.

Near-fault observations can provide insights into the physical process interaction between fault slip activation, fluid presence/migration and seismicity production, processes acting at different timescales that generate large earthquakes.The DETECT experiment aims at exploiting very dense seismic networks deployed across a segmented faults system to foster the development of scientific integrated methodologies for monitoring and imaging the faults behavior during the inter-seismic phase. Target of the monitoring is to: detect and track space-time trends of different source parameters that could be related to a preparation process leading to a larger earthquake; investigate the frictional and stress states of the fault segments to anticipate the characteristics of the future large earthquake (e.g., hypocenter but also future large seismic energy release locations); analyze the interactions between the different fault segments to model/anticipate potential cascade effects.The DETECT experiment is carried out in the Irpinia area (southern Italy), one of the regions in Italy and Europe showing the highest seismic hazard. Since august 2021, a constellation of 20 seismic arrays, for a total of 200 seismic stations (20 broad-band sensors and 180 short-periods), has been installed over the fault segments responsible for the Ms 6.9, 1980 Irpinia earthquake, the strongest and most destructive seismic event of the last half-century in southern Italy.DETECT results from a joint effort of local Universities, National and International Research Institutes. A novel and crucial aspect is that, differently from most studies concerning intra-plate earthquakes which are usually carried out after that large magnitude earthquakes have occurred, DETECT aims to put us in advantageous position and to unveil the preparatory process that generate large earthquakes and anticipate the role of the segmentation by studying at one time different fault segments, some of which are in rather late stage of their seismic cycle.With this contribution, we aim to present the DETECT experiment, the preliminary results and foster additional cooperation including complementary expertise to further enrich the partnership.

This data publication provides a collection of ground-motion simulation output for potential future earthquakes in the Rhine Graben area, Germany. Such data can be used as input for other engineering calculations, such as dynamic response history analysis. The earthquake sources used for the simulation consist of a stochastic catalog. They were generated using the German national seismic hazard model, the event-set calculator in the OpenQuake engine (Pagani et al., 2014), and considering both areal seismic source and known tectonic faults in the area as seismic sources in the analyses (branch C in Grünthal et al.,2018). The generated events represent a possible realization of the seismicity in the area within 10,000 years with peak ground accelerations greater than 0.02g. To build the simulation database, ground-motion simulations are performed using these potential future earthquakes from stochastic catalog and adopting the simulation method of Graves and Pitarka (2010, 2015) implemented in the Southern California Earthquake Center (SCEC) broadband platform (BBP), which is tailored for use in the Rhine Graben, as discussed in Razafindrakoto et al. (2021). Here, the approach used to simulate the ground-motion is only briefly discussed; a more accurate description is given in Razafindrakoto et al. (2021), discussing the calibration and validation of ground-motion in the Rhine Graben area. Accordingly, the provided data consists of a simulation output of 284 scenario earthquakes at 76 virtual stations in the Rhine Graben area. The files provided here include the earthquake source, station information, and the simulation results in terms of time histories for individual simulation in one zip file, and a flatfile that combines various intensity measures for all sources and stations.

The data publication contains a dataset for fast assessment of earthquakes based on seismic waveforms. The dataset encompasses Northern Chile. Due to the large scale of the dataset, it is intended for use in machine learning. A similar dataset for chile has been published as Münchmeyer et al. (2020). A similar dataset for Japan can be obtained using the scripts at https://github.com/yetinam/TEAM The datasets are provided as a hdf5-file (Folk et al. 2011), a hierachical file format. Source code for reading and processing the data is available at https://github.com/yetinam/TEAM. The hdf5-file contains the two groups “metadata” and “data” that are described below. These groups are the hdf5-analog of folders in a file system.

Irpinia seismic Array is part of the DEnse mulTi-paramEtriC observations and 4D high resoluTion imaging (DETECT) project focused on the acquisition of a unique multiparametric dataset and fosters collaboration among various institutions. The University of Naples Federico II (UniNa) and the German Research Centre for Geosciences (GFZ) are leading this effort carried out in collaboration with various local institutions and supported by the local municipalities. The DETECT project aims at exploiting very dense seismic networks deployed across a segmented fault system (Irpinia and Pergola-Melandro) to foster the development of scientific integrated methodologies for monitoring and imaging the fault behavior during the inter-seismic phase. The Irpinia seismic Array consists of a dense constellation of seismic antennas using more than 200 seismic stations deployed for one year. Each seismic antenna, with maximum aperture of ~2 km, uses one broad-band sensor, one short period sensor with 1 Hz and 8 with 4.5 Hz natural frequency. The antennas are deployed above and near the fault segments that generated during the last centuries many strong earthquakes in the southern Apennines. Waveform data are available from the GEOFON data centre, under network code ZK.

Location and magnitude are the primary information released by any seismological observatory to characterize an earthquake. Nowadays, the size of large enough earthquakes are routinely measured in terms of released seismic moment (moment magnitude, Mw). Whereas events with Mw above about 5.5 populate seismological archives connected to global monitoring networks, the moment magnitude of smaller events require the analysis of regional and local dense networks, or the establishment of empirical relationships to convert other magnitude scales into Mw (e.g., local magnitude to moment magnitude conversions). Since Mw is constructed over a physical parameter, it does not saturate. Moreover, being the seismic moment connected to tectonic features such as fault area and the average slip, Mw became the reference magnitude for seismic hazard studies. Although Mw accomplishes perfectly the task of characterizing the earthquake size, it does not provide the most complete view about the earthquake strength since Mw is insensitive to changes in the rupture dynamics. An assessment of the amount of the seismic energy released by an event (energy magnitude Me) is allowing to complement Mw with a measure of the earthquake size more suitable to evaluate the earthquake shaking potential.Aiming at introducing soon a new real-time service providing Me for major earthquakes we are presenting in this study the results of benchmark tests against the procedure proposed by Di Giacomo et al., in 2008 [1] as well as the analysis performed on a larger data set including all major events available in the GEOFON catalogue with a published moment magnitude since 2011. The initial procedure has been translated to a python code within the Stream2segment package [2] and leveraging on EIDA and IRIS data services, more than 2000 station for ~5000 events since 2011 have been downloaded and processed. The large data set used and the real-time application pose new challenges, among them the teleseismic distance, the strongly unbalanced network and the real-time data flow making the data set used dynamic. We present and discuss here the effects of these complications and how we are tackling them towards the implementation of new service at GFZ computing Me in real-time. [1] Di Giacomo, D., Grosser, H., Parolai, S., Bormann, P., and Wang, R. (2008), Rapid determination of Me for strong to great shallow earthquakes, Geophys. Res. Lett., 35, L10308, doi:10.1029/2008GL033505.[2] Riccardo Zaccarelli, Dino Bindi, Angelo Strollo, Javier Quinteros, Fabrice Cotton; Stream2segment: An Open‐Source Tool for Downloading, Processing, and Visualizing Massive Event‐Based Seismic Waveform Datasets. Seismological Research Letters ; 90 (5): 2028–2038. doi: https://doi.org/10.1785/0220180314

The data publication contains a dataset for fast assessment of earthquakes and early warning based on seismic waveforms. The dataset encompasses Italy and surrounding refions. Due to the large scale of the dataset, it is intended for use in machine learning. A similar dataset for Japan, with the same specifications as the one provided in this data publications, can be obtained using the scripts at https://github.com/yetinam/TEAM

Onsite earthquake early warning (EEW) systems exploit predictive models relating fea- tures extracted over the P-wave window to S-wave ground-motion parameters. These models are usually calibrated considering recordings from multiple stations and combining datasets from different regions under the ergodic assumption. Here, we show that the local-site conditions can play a significant role in determining the performance of onsite EEW predictive models in terms of rates of false or missed alerts. Interestingly, if partially nonergodic models are implemented, as done in probabilistic seismic hazard analysis, the negative impact of local-site amplifications can be mitigated. We explore the influence of site effects for onsite EEW predictive models calibrated between the peak displacement (Pd ) and integral squared velocity (Iv2) measured over a 3 s P-wave window, and the accel- eration response spectra (RSA) at nine different periods T (T 0:1, 0.15, 0.2, 0.3, 0.5, 0.75, 1.0, 1.5, and 2.0 s). We consider 58 earthquakes with magnitudes between Mw 3.7 and 6.5, belonging to the 2016–2017 central Italy seismic sequence that have been recorded by 100 accelerometer stations at hypocentral distances lesser than 150 km. We implement a mixed-effects regression analysis to explore the variability of the ground motion in terms of RSA predicted at different sites by considering two different group levels: in the first, each station is considered separately; in the second, we consider the Eurocode 8 (EC8, 2004) soil classification. Considering a probabilistic alert decision module applied to data from two selected stations, we show that the predictive models including site effects provide more reliable alerts, reducing the false alarms from 2.6% to 0.53% and the missed alarms from 10.1% to 4.8%. The residuals analysis shows that including a site-specific random effect in the predictive model contributes to reducing the apparent aleatory variability, whereas grouping data by EC8 classification does not provide significant benefit for EEW purposes.

In Bindi et al. (2019) a harmonized local magnitude scale across Europe has been derived using data disseminated by network operators through the European Integrated Data Archive (EIDA). This data set contains regionalized non-parametric attenuation tables, attenuation corrections to the parametric model and station corrections for both non-parametric and parametric models for more than 2000 stations in Europe. Regionalization has been performed considering six different regions covering Europe and the polygons defining them are also provided. Data are subject to updates that can be triggered by the availability of new and substantial input data (reviewed earthquake catalogues and/or new waveforms). Each update will be released with a new version of the data. The data are provided in ASCII format (.csv).