The 2016, October 26, Central Italy Earthquake Origin Time 17:10:36 UTC, M L(ISNet) =5.3; M W(ISNet) =5.6

Transcription

1 The 2016, October 26, Central Italy Earthquake Origin Time 17:10:36 UTC, M L(ISNet) =5.3; M W(ISNet) =5.6 RISSC-Lab: Laboratorio di RIcerca in Sismologia Sperimentale e Computazionale

2 The event as seen from ISNet We present the results related to the Central Italy earthquake, as seen from ISNet (Irpinia Seismic Network). They concern the magnitude, source parameters, focal mechanism and ground shaking estimations. They are automatically obtained and then revised from RISSC-Lab team. 237 Km The Closest station is BENI (237 km) The Farthest one is MRN3 (348 km)

3 Waveforms Vertical component of accelerometers and velocimeters from ISNet ordered as a function of distance (distance is increasing from bottom to top).

4 Acceleration [cm/s 2 ] Acceleration [cm/s 2 ] Three-component accelerograms UD UD NS NS PGA = 0.31 cm/s PGA = 0.08 cm/s EW EW Time [s] Time [s] Three-component acceleration records at the closest (BENI) and farthest (MRN3) stations

5 Magnitude and source parameters ML Mw M 0 (Nm) fc (Hz) Radius (km) Δσ (MPa) (3.2 ±1.1) 0.43± ± ± ± ± 3 3 * Procedures for source parameters computation are described at the end of the presentation The corner frequency retrieved from spectral inversion indicates a rupture size of about 5.2 km, which is compatible with scaling laws of Wells and Coppersmith (1994). Observed and inverted displacement spectra at a subset of stations.

6 Focal Mechanism The two solutions are: Str/dip/Slp: 124/53/40; 8/60/136 Mw = 5.29 Depth = 9 km Reduced Var.= sec

7 PGV across the network The largest value of PGV ( cm/s) at ISNet was recorded at the station BENI. The PGV variability across the network is about a factor 7. PGV distribution across the network

8 PGA across the network The largest value of PGA (0.04 %g, 0.36 cm/s 2 ) was recorded at the station BENI (epicentral distance 237 km). The PGA variability across the network is about a factor 6. PGA distribution across the network

9 PGA and PGV M = 5.3 M = 5.3 Observed PGA and PGV compared with Bindi et al. (2011), for M = 5.3

10 The 2016, October 26, Central Italy Earthquake Origin Time 19:18:05 UTC, M L(ISNet) =5.8; M W(ISNet) =6.0 RISSC-Lab: Laboratorio di RIcerca in Sismologia Sperimentale e Computazionale

11 The event as seen from ISNet We present the results related to the Central Italy earthquake, as seen from ISNet (Irpinia Seismic Network). They concern the magnitude, source parameters, focal mechanism and ground shaking estimations. They are automatically obtained and then revised from RISSC-Lab team. 241 Km The Closest station is BENI (241 km) The Farthest one is MRN3 (352 km)

12 Waveforms Vertical component of accelerometers and velocimeters from ISNet ordered as function of distance (distance is increasing from bottom to top).

13 Acceleration [cm/s 2 ] Acceleration [cm/s 2 ] Three-component accelerograms UD 0.10 UD NS 0.10 NS PGA = 0.16 cm/s EW 0.10 EW PGA = 0.91 cm/s Time [s] Time [s] Three-component acceleration records at the closest (BENI) and farthest (MRN3) stations

14 Latitude Array seismic location Longitude INGV loc. Array loc. We show the comparison between event location obtained with array tecnique and the reference INGV location: Array Lat = Lon = 13.3 INGV Lat = Lon = The distance between the two locations is about 14 km. Azimuth (angle measured in clockwise direction from the north line) and distance between the event and VDS station (red triangle) are: Array Azimut= 325 Distance = 304 km INGV Azimut= 322 Distance = 308 km

15 Magnitude and source parameters ML Mw M 0 (Nm) fc (Hz) Radius (km) Δσ (MPa) (1.3 ±0.4) 0.38± ± ± ± ± 10 4 * Procedures for source parameters computation are described at the end of the presentation The corner frequency retrieved from spectral inversion indicates a rupture size smaller (~6 km) than expected for a M=6.0 event and observed from aftershock distribution. Observed and inverted displacement spectra at a subset of stations. This can be due to a directivity effect and/or to the occurrence of largest slip in a smaller size asperity. The stress drop can be hence overestimated

16 Focal Mechanism The two solutions are: Str/dip/Slp: 132/53/66; 349/43/119 Mw = 5.84 Depth = 8 km Reduced Var.= sec

17 PGV across the network The largest value of PGV (0.17 cm/s) at ISNet was recorded at the station BENI. The PGV variability across the network is about a factor 8. PGV distribution across the network

18 PGA across the network The largest value of PGA (0.11 %g, 1.05 cm/s 2 ) was recorded at the station BENI (epicentral distance 241 km). The PGA variability across the network is about a factor 11. PGA distribution across the network

19 PGA and PGV M = 5.8 M = 5.8 Observed PGA and PGV compared with Bindi et al. (2011), for M = 5.8

20 Earthquake Early Warning at ISNet PRESTo EW system SAVE EW system Two EW systems run at ISNet, the regional system PRESTo, which analyzes the data of the whole network and the onsite system SAVE, that operates on 7 stations. Details about the systems can be found in Appendix. Both systems did not associate the events. The P-wave arrivals were emergent and they did not meet the picking criteria. Hence the systems did not issue an alarm for the Irpinia region. Nevertheless SAVE estimated an intensity corresponding to weak shaking and no damage for the area.

21 Technical details

22 Array seismic location technique The array techniques used in this work is based on a combination between the plane wave approximation (PWF, Del Pezzo & Giudicepietro 2001), which is realistic for epicentral distances greater than four or five times the array aperture, and circular wave approximation (CWF, Almendros et al. 1999), which becomes necessary for closer distances. The PWF method allows the estimation of azimuth and slowness of an incoming wave using the ISNet network as a seismic array and using as data the first arrival times of P-wave. The CWF method, using as initial solution the estimates obtained with PWF, allows to explore the distance parameter trough the downhill simplex inversion technique (Nelder & Mead 1965). The array location is obtained using first arrival times, manually picked, of P- wave recorded at 11 station of ISNet network.

23 Parameter computation Source parameters are computed fixing the location of the event at the epicenter as defined by the INGV Ml is computed using 13 stations and the law provided by of Bobbio et al. (2009) Mw and fc are obtained by the inversion of the displacement spectra at 7 stations (Zollo et al., 2014). Source radius and stress drop are computed from seismic moment and corner frequency using the Brune model. Focal mechanism is computed from inversion of the moment tensor using the TDMT program (Dreger et al., 2003) and 4 stations from ISNet, in the frequency band Hz.

24 PRESTo Early Warning System PRESTo (PRobabilistic and Evolutionary early warning SysTem) (Satriano e al., 2010) is a software platform for the immediate detection of seismic events using the real-time data from a seismic network. This approach to Early Warning is called regional or network-based. It is aimed at the rapid estimation of the earthquake source parameters (location, origin time, magnitude) and, through Ground Motion Prediction Equations, of the maximum ground shaking in the epicentral area (PGA, PGV, I MM ). The system is being tested at the RISSC-Lab (RIcerca in Sismologia Sperimentale e Computazionale), where it receives in real-time and in continuous the ground acceleration measured at the ISNet stations (Irpinia Seismic Network), managed by AMRA scarl (Analisi e Monitoraggio Rischio Ambientale). In case of a potentially damaging event in the proximity of the network, the system is able to estimate the location and the magnitude within a few seconds after the origin time and to send these data to remote sites to be protected during the propagation time of the seismic waves. The alert messages also contain the estimation of the maximum ground velocity and maximum ground acceleration for each specific site and the remaining seconds before the arrival of the first S-wave. These estimations are continuously updated together with their uncertainties and can enable to alert target structures up to tens of seconds before the arrival of the energetic waves there, depending on the distance to the source.

25 SAVE Early Warning System SAVE (on-site Alert level) is a software that exploits the real-time ground acceleration measured by a single sensor, installed at a site to protect. During an earthquake it can rapidly estimate the damaging effects at the site, before they occur. This approach to Earthquake Early warning, is called single-station or on-site. It is complementary to the regional based approach used by PRESTo, where the target site alert is instead based on the signal recorded near the source by several stations, rather than at the target itself by a single station. SAVE uses the first few seconds of the P-waves, recorded at the site to protect, to estimate, ahead of time, the Intensity that will be measured at the site. This parameter is of primary concern to initiate safety measures. The estimate is particularly robust since it is based on the actual signal measured at the target. The software can also estimate the earthquake magnitude and the distance of the source to the site to protect, but only when the quality of the recorded P-wave signal is good enough (i.e. larger magnitude and/or smaller distance to the earthquake source). SAVE is being tested in real-time on several stations of the ISNet network, each serving as a target site to protect, in parallel with the regional system PRESTo. SAVE is particularly useful for sites located too near the epicenter for the regional EEWS to be fast enough to provide a useful alert (the so-called blind zone). In these cases the on-site system is instead able to provide a robust warning even before the arrival of the strong shaking waves.

26 References Almendros, J., Ibáñez, J.M., Alguacil, G. and Del Pezzo, E., Array analysis using circular-wave-front geometry: an application to locate the nearby seismo-volcanic source. Geophysical Journal International, 136(1), pp Bobbio A., Vassallo M., Festa G. A local magnitude scale for southern Italy, Bullettin of of Seismological Society of America,Vol. 99, No.4, , 2009 doi: / Brune J.N., (1970). Tectonic stress and the spectra of seismic shear waves from earthquakes. J. Geophys. Res. 75, S. Colombelli, A. Emolo and A. Zollo. A Duration Magnitude Scale for the Irpinia Seismic Network, Southern Italy Seismological Research Letters, January/February 2014, v. 85, p , doi: / Del Pezzo, E. and Giudicepietro, F., Plane wave fitting method for a plane, small aperture, short period seismic array: a MATHCAD program. Computers & geosciences, 28(1), pp Dreger, D. S. (2003) TDMT_INV: Time domain seismic moment tensor inversion. International Geophysics, 81, Nelder, J.A. and Mead, R., A simplex method for function minimization. The computer journal, 7(4), pp Satriano, C., Elia L., Martino C., Lancieri M., Zollo A., Iannaccone, G. PRESTo, the earthquake earlywarning system for Southern Italy: Concepts, capabilities and future perspectives Soil Dynamics and Earthquake Engineering, 2010, doi: /j.soildyn Scognamiglio, L., Tinti, E., & Michelini, A. (2009). Real-time determination of seismic moment tensor for the Italian region. Bulletin of the Seismological Society of America, 99(4), Zollo A., Orefice A., Convertito V. (2014). Source Parameter Scaling and Radiation Efficiency of Microearthquakes Along the Irpinia Fault Zone in Southern Apennines, Italy, submitted to J. Geophys. Res.

27 Useful links: ISNet Bulletin ISNet RISSC-lab