BOZZONI_corretto_Layout 6 609 Preliminary results of ground-motion characteristics Francesca Bozzoni1,*, Carlo G. Lai1,2, Laura Scandella1 1 European Centre for Training and Research in Earthquake Engineering (EUCENTRE), Pavia, Italy 2 Università di Pavia, Dipartimento di Ingegneria Civile ed Architettura, Pavia, Italy ANNALS OF GEOPHYSICS, 55, 4, 2012; doi: 10.4401/ag-6121 1. Introduction On May 20, 2012, a MW 5.9 earthquake struck the Emilia-Romagna Region in northern Italy (Figure 1a), caus- ing severe shaking throughout a relatively large portion of the Po Plain territory. The epicenter of this May 20, 2012, event has the coordinates 44.89˚ (latitude) and 11.23˚ (lon- gitude). The depth of the hypocenter has been estimated at 6.3 km. The focal mechanism is of the thrust type. On May 29, 2012, a MW 5.65 shock hit the region, causing further damage and fatalities. Figure 1b shows the events with magnitudes >4.5 (Figure 1b, red dots) of this sequence of May- June 2012 (data retrieved from http://iside.rm.ingv.it/iside/ standard/index.jsp). Figure 1b also shows the most important earthquakes that have occurred in the northern parts of the Emilia-Romagna Region over the latest 500 years (Figure 1b, green dots). The data were retrieved from the Italian earth- quake catalogue CPTI11 (http://emidius.mi.ingv.it/CPTI11/) of the Istituto Nazionale di Geofisica e Vulcanologia (INGV; National Institute of Geophysics and Volcanology). The Fer- rara territory was hit by a M 5.5 earthquake on November 17, 1570. On July 11, 1987, a M 5.4 event struck the Po Plain area close to Bologna and Ferrara. Other historical earthquakes with magnitudes up to 6 have occurred in the southern part of the Emilia-Romagna Region, close to the Apennine chain. The area struck by the Emilia May-June 2012 sequence is located south of the Po Plain, in the foreland basin of two mountain belts constituted by the Alps and the northern Apennine chains. Under thick clastic sedimentary fills along the northern and southern margins of the Po Plain, complex systems of thrust sheets and tectonic structures are buried. Due to the fast sedimentation rates and comparatively low tectonic rates, the thrusts are generally buried and there is little surface evidence of their activity [Toscani et al. 2009]. 2. Shake maps Shake maps provide a first estimate of the spatial distri- bution of ground motion. Data from the recording stations of the Italian Strong Motion Network (RAN) and the Strong Motion Network of Northern Italy (RAIS) have been used for this purpose. The maximum peak ground acceleration (PGA) was recorded at the station of Mirandola (MRN) (lo- cated at ca. 13.4 km from the epicenter), and it is associated to the vertical component, which reached 0.31 g. Figure 2 shows a shake map computed by interpolating the recorded PGA at the RAN stations using the inverse distance weighted algorithm. The stations are located at sites classified as cate- gory C according to the soil classification system prescribed by the Italian Building Code [NTC 2008] which is similar to that of Eurocode 8 Part 1 [EN 1998-1, 2005]. From Figure 2 it is worth noting that the pattern of the contour lines of the PGA is nonsymmetric with respect to the epicenter. This is due to the small number of stations used to perform the interpolation (the station closest to the epicenter is MRN). The focal mechanism might have also had a role in this asymmetry. It is important to note that the distance between the epicenter and the closest station influ- ences the quality of the PGA estimation in the area. Ground-motion parameters from the recordings pro- vided by the RAN and RAIS seismic networks were com- pared with the estimates of two recent ground-motion prediction equations (GMPEs); namely those of Cauzzi and Faccioli [2008] and Bindi et al. [2011]. The preliminarily pre- dictions were made for Soil Category C. Figure 3 and Figure 4 show the horizontal PGA predicted by the selected GMPEs. Particularly in the case of the Cauzzi and Faccioli [2008] GMPE, the predicted values are very high in the epicentral area. The predicted PGA is in good agreement with the recordings from the RAN and RAIS stations. 3. Recordings and response spectra at Mirandola station The closest station to the epicenter of the May 20, 2012, event is MRN, which belongs to the RAN network (managed by the Italian Department of Civil Protection). The epicen- tral distance of this station is ca. 13.4 km. Time histories recorded at the MRN station have been analyzed to obtain acceleration, velocity, and displacement response spectra. Article history Received July 22, 2012; accepted August 20, 2012. Subject classification: Seismology - Ground motion. 2012 EMILIA EARTHQUAKES The recorded signals, which are the horizontal South-North (SN), East-West (EW) and vertical (UP) components, were processed for standard baseline correction, band-pass filtered from 0.05 Hz to 50 Hz, tapered, and linearly de-trended in velocity. Table 1 shows the peak ground-motion parameters, namely the PGA, the peak ground velocity (PGV), and the peak ground displacement (PGD), for each component recorded at the MRN station. The acceleration response spectra computed using the recordings of the MRN station are shown in Figure 5a (hor- izontal components) and in Figure 5b (vertical component). The vertical component of the spectral acceleration has the peak at a lower period (0.06 s), compared with the horizon- tal components (SN: 0.17 s; EW: 0.31 s). Furthermore, the peak values of the spectral acceleration for the vertical com- ponent are concentrated in a narrow band of periods from 0.02 to 0.12 s, while the peaks of the spectral acceleration for the horizontal components occur at a band with higher pe- riods of up to 0.5 s (EW) and 1.3 s (SN). The acceleration response spectra computed using the recordings from the MRN station were also compared with the predictions of the GMPE developed by Bindi et al. [2011] for the horizontal component and for Soil Category C (Figure 6a). The mean predictions of the GMPE are satis- factory for periods <0.17 s, whereas for larger periods, the GMPE strongly underestimates the severity of the record. The displacement response spectra computed using the recordings from the MRN station were compared with the spectra predicted by the GMPE developed by Cauzzi and Faccioli [2008] for the horizontal component and for Soil Category C (Figure 6b). In this case, the GMPE severely un- derestimates the spectra calculated from the records for all of the structural periods. The records from the MRN station were used to plot the particle orbit (hodogram) described by the waveform. The hodogram was plotted in the plane of Rayleigh wave propa- BOZZONI ET AL. 610 PGA (g) PGV (m/s) PGD (m) SN 0.264 0.463 0.105 EW 0.262 0.300 0.081 UP 0.310 0.059 0.018 Figure 1. (a) Location in the Emilia-Romagna Region of the May 20, 2012, earthquake. (b) Distribution of the major historical earthquakes as green dots (http://emidius.mi.ingv.it/CPTI11/), and of the M >4.5 events of the 2012 sequence as red dots (http:/iside.rm.ingv.it/). Table 1. Peak ground-motion parameters for each component recorded at the MRN station. PGA, peak ground acceleration; PGV, peak ground ve- locity; PGD, peak ground displacement). a) b) 611 GROUND-MOTION CHARACTERISTICS Figure 2. Shake map obtained from the interpolation of the PGA recorded by the RAN seismic network. Ground conditions for the seismic stations are similar, and they are approximately associated to the Soil Category C, according to the ground classification system of the Italian Building Code [NTC 2008], which is similar to that of Eurocode 8 Part 1 [EN 1998-1, 2005]. Figure 3. Median values of the PGA (horizontal component) predicted by the GMPE of Cauzzi and Faccioli [2008] for Soil Category C, plotted on top of the recorded values of the PGA at selected stations of the RAN and RAIS strong-motion networks. Figure 4. Median values of the PGA (horizontal component) predicted by the GMPE of Bindi et al. [2011] for Soil Category C, plotted on top of the recorded values of the PGA at selected stations of the RAN and RAIS strong-motion networks. gation. Inspection of the MRN recordings shows that the backscattered P-wave and SV-wave are not well time-separated from the Rayleigh wave arrival, given that the MRN station is very close to the epicenter. Figure 7 shows the radial and ver- tical displacement traces obtained through filtering the signals with band-passes between 0.1 Hz and 0.5 Hz (from 2 s to 10 s). The motion is characteristically retrograde elliptical. 4. Preliminary ground-response analysis at Mirandola The availability of results from geotechnical field tests performed at a few sites in the Municipality of Mirandola prior to the earthquake of May 2012 have allowed preliminary as- sessment of the ground response closest to the MRN station. The corresponding data were retrieved from the WebGIS por- tal of the Emilia-Romagna Region (https://territorio.regione. emilia-romagna.it/cartografia/cartografia-sgss). Data from five seismic tests using multichannel analyses of surface waves that were performed during a 2011 geophysical campaign were used to construct a geotechnical model of the subsoil at Mirandola, which was successively adopted to perform ground-response analyses. Geological information about the soil deposits at the MRN site was retrieved at the portal BOZZONI ET AL. 612 Figure 5. Mirandola station recordings. (a) Comparison of the horizontal (SN, EW) acceleration response spectra from the recordings with the spectra from the Italian Building Code [NTC 2008], computed for Soil Categories C and D for 475-year and 975-year return periods. (b) Comparison of the vertical (UP) acceler- ation response spectra from the recordings with the spectra from the Italian Building Code [NTC 2008] computed for 475-year and 975-year return periods. Figure 6. Mirandola station recordings. (a) Comparison of the horizontal (SN, blue line; EW, black line) acceleration response spectra from the record- ings with the Bindi et al. [2011] GMPE for ground category C. (b) Comparison of the horizontal (SN, blue line, EW, black line) displacement response spectra from the recordings with the Cauzzi and Faccioli [2008] GMPE for Soil Category C. a) b) a) b) Figure 7. Hodogram from the Mirandola station recordings: the radial and vertical displacement traces. 613 http://itaca.mi.ingv.it/ItacaNet/. Table 2 gives the mean val- ues of the geotechnical parameters and the corresponding uncertainties assumed for the adopted subsoil model. One-dimensional (1D) linear-equivalent, fully stochastic site-response analysis was performed for the MRN station using the methodology described in Rota et al. [2011], which takes into account the uncertainties associated with the ge- otechnical model parameters. The stochastic analysis was car- ried out using Monte Carlo simulations associated with the Latin Hypercube sampling technique. Randomly generated geotechnical parameters varying within correctly defined probability distributions were assumed to calculate the seis- mic response of 100 deterministic realizations of the geot- echnical model. The uncertainty adopted for the values of VS is shown in Figure 8a where the red line denotes the mean of 100 VS profiles with layers of varying thicknesses. For the defi- nition of the depth of the bedrock and of its stiffness, the data available at the website of the Emilia-Romagna Region were used (http://ambiente.regione.emilia-romagna.it). The variability of the seismic input was taken into ac- count by considering an appropriate set of seismo- and spec- trum-compatible natural records. Specifically, a set of seven accelerograms that were recorded from actual earthquakes were downloaded for the area under investigation, from the portal: http://www.eucentre.it/seismhome.html. The se- lected accelerograms have magnitudes that range from 6 to 6.87, and epicentral distances that vary from 11 km to 102 km. Their scaling factors range from 0.36 to 2.61, with a mean of 1.60. The signals were recorded under outcropping rock conditions, and they are spectrum-compatible, on aver- age, to the Italian code-based spectrum [NTC 2008] that refers to the 475-year return period. Indeed, although it is still premature to make definitive conclusions, a lot of seismo- logical evidence appears to suggest that the May 20, 2012, earthquake might correspond to this return period. Figure 8b shows the mean acceleration response spec- trum ±one standard deviation, which was calculated using the linear-equivalent, fully stochastic, 1D ground response GROUND-MOTION CHARACTERISTICS Figure 8. Stochastic ground-response analyses at Mirandola. (a) One hundred random VS profiles generated using the Latin Hypercube sampling tech- nique. Red line, mean profile. (b) Mean acceleration response spectrum with the associated scatter computed using the VS profiles of (a). The seismic input comprised seven actual records compatible with the Italian code-based spectrum that refers to a 475-year return period. The computed spectra are com- pared to the SN and EW spectra calculated from the MRN station recordings. The spectrum code [NTC 2008] for Soil Category A is also shown. Layer Soil type Thickness (m) Uncertainty (%) VS (m/s) Uncertainty (%) t (kN/m3) Uncertainty (%) 1 Sand 6.6 67 150 12 18 6 2 Sand 7.8 68 221 33 18 6 3 Sand 9.7 67 266 19 18 6 4 Sand 7.1 44 312 18 18 6 5 Sandy clay and clayey sand 10.0 50 379 20 19 6 6 Sandy clay and clayey sand 15.0 50 480 25 19 6 7 Sandy clay and clayey sand 15.0 50 600 20 19 6 8 Sandy clay and clayey sand 35.0 40 700 20 19 6 - Bedrock - - 800 20 20 6 Table 2. Means of the geotechnical parameters and the corresponding uncertainties assumed for the adopted subsoil model. a) b) analyses described above. This spectrum is compared to the response spectra calculated from the SN, EW recordings of the MRN station for the May 20, 2012, event. Figure 8b also shows the code-based spectrum for Soil Category A [NTC 2008]. The mean PGA computed from ground-response analysis turned out to be 0.269 g, which is in excellent agree- ment with the recorded PGA (see Table 1). However the peaks of the MRN spectra are underestimated by the mean spectrum. Better agreement appears to be obtained by con- sidering the mean spectrum plus one standard deviation, es- pecially in the case of the EW component, although only for periods up to ca. 1 s. For periods >1 s, the spectral accelera- tions of the MRN recordings, and in particular those of the SN component, exceed those of the computed spectra. Possible explanations for this difference might be near- fault effects, as the MRN station lies only 13 km away from the epicenter, and also that the May 20, 2012, earthquake was a shallow event (with a focal depth of about 6.3 km). Possible explanations for such differences at large periods might also be non-linear effects (liquefaction), which might have oc- curred at the recording station. Further investigation is how- ever needed to substantiate these preliminary statements. Ground-response analyses focused on the propagation of S-waves only, as they are the most relevant for engineering applications. Future studies can investigate the role, if any, of ground amplification in the propagation of P-waves. 5. Concluding remarks The preliminary results of our ground-motion calcula- tions from the recordings of the May 20, 2012, M 5.9 Emilia earthquake have been presented here. The response spectra computed from the accelerograms recorded at the MRN sta- tion, as the closest to the epicenter, have been compared with the spectra predicted by two recent GMPEs and the Italian Building Code [NTC 2008]. A first attempt evaluation of the amplification effects at Mirandola has been carried out through linear-equivalent, fully stochastic ground-response analyses based on the results from geotechnical field tests performed close to the MRN station and using a set of seven actual records that are compatible to the Italian code-based spectrum that refers to a 475-year return period. Investiga- tions are currently underway to determine whether linear- equivalent analyses are admissible in Mirandola, in light of possible strongly nonlinear effects (liquefaction) that might have occurred at the recording station. The results of the analyses show that the mean com- puted PGA is in excellent agreement with the recorded PGA at the MRN station (SN, EW). However, the peaks of the MRN spectra are underestimated by the mean spectrum. Better agreement appears to be obtained by considering the mean spectrum plus one standard deviation, although only for periods up to ca. 1 s. For periods >1 s, the spectral accel- erations of the MRN recordings, and especially of the SN component, exceed those of the computed spectra. Possible explanations for this difference include near-fault effects (the MRN station lies only about 13 km away from the epicenter) and that the May 20, 2012, earthquake was a shallow event (focal depth of 6.3 km). Acknowledgements. The authors would like to acknowledge the sup- port of the Department of Civil Protection of the Italian Government, and that of the Administration of the Emilia-Romagna Region. A word of grat- itude goes to Dr. Maria-Daphne Mangriotis, for her valuable suggestions. References Bindi, D., F. Pacor, L. Luzi, R. Puglia, M. Massa, G. Ameri and R. Paolucci (2011). Ground motion prediction equa- tions derived from the Italian strong motion database, B. Earthq. Eng., 9, 1899-1920. Cauzzi, C., and E. Faccioli (2008). Broadband (0.05 to 20 s) prediction of displacement response spectra based on worldwide digital records, J. Seismol., 3; doi:10.1007/s10 950-008-9098-y. EN 1998-1 (2005). Eurocode 8 - Design of Structures for Earthquake Resistance - Part 1: General rules, seismic ac- tions and rules for buildings, EN 1998-1, CEN, Brussels. NTC (2008). Norme tecniche per le Costruzioni, D.M. 14.1.2008 (Italian Building Code). Rota, M., C.G. Lai and C.L. Strobbia (2011). Stochastic 1D site-response analysis at a site in central Italy, Soil Dyn. Earthq. Eng., 31, 626-639. Toscani, G., P. Burrato, D. Di Bucci, S. Seno and G. Valensise (2009). Plio-Quaternary tectonic evolution of the north- ern Apennines thrust fronts (Bologna-Ferrara section, Italy): Seismotectonic implications, B. Soc. Geol. Ital., 128, 605-613. *Corresponding author: Francesca Bozzoni, European Centre for Training and Research in Earthquake Engineering (EUCENTRE), Pavia, Italy; email: francesca.bozzoni@eucentre.it. © 2012 by the Istituto Nazionale di Geofisica e Vulcanologia. All rights reserved. BOZZONI ET AL. 614 << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles false /AutoRotatePages /None /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.3 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams true /MaxSubsetPct 100 /Optimize false /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile (None) /AlwaysEmbed [ true /AndaleMono /Apple-Chancery /Arial-Black /Arial-BoldItalicMT /Arial-BoldMT /Arial-ItalicMT /ArialMT /CapitalsRegular /Charcoal /Chicago /ComicSansMS /ComicSansMS-Bold /Courier /Courier-Bold /CourierNewPS-BoldItalicMT /CourierNewPS-BoldMT /CourierNewPS-ItalicMT /CourierNewPSMT /GadgetRegular /Geneva /Georgia /Georgia-Bold /Georgia-BoldItalic /Georgia-Italic /Helvetica /Helvetica-Bold /HelveticaInserat-Roman /HoeflerText-Black /HoeflerText-BlackItalic /HoeflerText-Italic /HoeflerText-Ornaments /HoeflerText-Regular /Impact /Monaco /NewYork /Palatino-Bold /Palatino-BoldItalic /Palatino-Italic /Palatino-Roman /SandRegular /Skia-Regular /Symbol /TechnoRegular /TextileRegular /Times-Bold /Times-BoldItalic /Times-Italic /Times-Roman /TimesNewRomanPS-BoldItalicMT /TimesNewRomanPS-BoldMT /TimesNewRomanPS-ItalicMT /TimesNewRomanPSMT /Trebuchet-BoldItalic /TrebuchetMS /TrebuchetMS-Bold /TrebuchetMS-Italic /Verdana /Verdana-Bold /Verdana-BoldItalic /Verdana-Italic /Webdings ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 150 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.10000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.10000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.08250 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /CreateJDFFile false /SyntheticBoldness 1.000000 /Description << /ENU (Use these settings to create PDF documents with higher image resolution for high quality pre-press printing. The PDF documents can be opened with Acrobat and Reader 5.0 and later. These settings require font embedding.) /JPN /FRA /DEU /PTB /DAN /NLD /ESP /SUO /NOR /SVE /KOR /CHS /CHT /ITA >> >> setdistillerparams << /HWResolution [2400 2400] /PageSize [595.000 842.000] >> setpagedevice