Site Effects Assessment Using Ambient Excitations sesame european Commission – Research General Directorate Project No. Evg1-ct-2000-00026 sesame report of the wp04

Download 131.01 Kb.
Date conversion18.07.2018
Size131.01 Kb.
1   2   3   4

Damage database acquisition

The damage database prepared in the framework the Umbria-Marche microzonation project (see the CD-ROM included in Marcellini and Tiberi, 2000) had the drawback of not being georeferenced and it lacked of synthetic definitions of non-structural and structural damage for each building as published in Dolce and Larotonda (2001).

The damage maps considered as the most representative for this experiment (Dolce e Larotonda, 2001) were consequently acquired in a digital raster format and georeferenced within the ArcViewTM GIS using a reference vectorial building theme provided by the Fabriano municipality. The damage data contained in the georeferenced images were manually digitized to obtain a point theme where each point have two indexes comprised between 0 and 3 for the description of non-structural and structural damage (see Figure 4).

Figure 5 EMS98 damage distribution in Fabriano.
The EMS 98 damage grade was computed by processing the point theme with a translation algorithm based on EMS 98 damage definition criteria (Grünthal, 1998). In this contest, the levels of structural and non-structural damage defined by Dolce and Larotonda were equalized to the first three levels of damage contained in the EMS 98 (e.g. EMS 98 slight non-structural damage corresponds to the light non-structural damage defined by Dolce and Larotonda, 2001). The point theme with the computed EMS 98 damage grade is represented in Figure 5.

Figure 6 Weak motion and noise point of measure in Fabriano. Short codes represent points where both weak motion and microtremor were measured while long names identify points where only microtremor measures were carried out.

Earthquake recordings and seismic noise data

With the specific aim of site effects assessment, a dense velocimetric network monitored 21 sites in the urban area of Fabriano for a period of about two months (from October to November, 1997). A detailed description of the deployment of the array, data acquisition and processing is contained in de Franco et al., 2000, whereas Franceschina, 2000, fully describes the obtained database of the selected events and noise samples. Tento et al., 2001, performed the site effects zonation of this area whose main steps are illustrated in the following.

The array was deployed mainly on the fluvio-lacustrine deposits that characterize the geology of the area investigated but for two stations that were installed on the outcrops of the Umbro-Marchigiana series to be taken as reference sites for the analyses (Figure 6).

The collected data allowed the creation of a database with about 1500 seismometric 3D recordings corresponding to 114 seismic events (70 events per site on the average) whose Magnitude ranges from ML 2.1 to 4.6 while their hypocentral distances are between 25 to 40 km and their back azimuth, with respect to the center of the array, are between 170 to 220 degrees.

Microtremor measurements were performed in the same sites of the seismological stations only in 5 cases whereas they are also available in 10 additional sites.

The data recorded evidenced a remarkable variability of site effects in the area investigated. For instance, the acceleration response spectra, 5% damping, of a ML 4.5 earthquake, generated at about 55 km, approximately the same source zone of the mainshocks, show a variability as high as 4 for PGA, 6.5 for T=0.1 s and 3.5 for T=0.5 s.

The basic elaborations of the velocimetric data, which are aimed to the construction of the site effects map, concern:

  • Analysis of the incidence angle of the first arrivals of P waves at the different stations with respect to that of the reference stations. The results are used to verify a qualitative correlation with the local geology of the monitored sites.

  • Computation of spectral ratios and receiver functions using a standard technique.

  • H/V spectral ratio of noise. Direct measurements of seismic noise were available at 15 sites, in these cases a standard procedure for the evaluation of H/V ratios was employed. For the remaining sites, corresponding to the other seismological stations, H/V ratio was estimated using, for each place, several pre-event windows opportunely selected from the available recordings. The ultimate result of this step is the map of the fundamental period of soil (T0). This map, redrawn with respect to the original one and reported in figure 10, was recomputed in terms of a triangulated irregular network (TIN) using a GIS software. We avoided the use of gridding algorithms, such as Inverse distance weighted methods or splines, because from our tests we observed a relevant variability on the distribution of maps especially in areas where the density of point of measure it is not satisfactory.

The above reported analyses led to construction of a map that characterize the examined area in terms of zones with overall homogeneous seismic behavior. The map, reported in figure 7, identifies these zones:

zone 0

Zone with amplification 1. Such zone is limited by the geologic boundaries of the Schlier and Bisciaro formations that can be considered as seismic bedrock on the base of geological evidences, receiver functions and H/V ratios of noise. Besides, the highest P wave incidence angles are observed in this zone.

zone 1

This zone, which approximately corresponds to zone “Borgo”, comprises three different kinds of lithotypes. It is characterized by spectral ratios which show the highest amplification experienced in the investigated area attaining values greater than 2.5 in a wide frequency band, from 2 to 8 Hz. Zone 1 is limited at South by the border of zone 0 and at East by the limit of the outcrop of the Gessosa Solfifera formation.

zone 2

This zone approximately encompasses the Fabriano historical center. It is characterized by spectral ratios with an amplification peak around 3 between 2 and 3 Hz. In this zone the incidence angle is comparable or greater than in zone 4. The isoline of T0 map corresponding to 0.5 s has been used to draw the Southern and Eastern limit of this zone.

zone 3

This zone includes two part located at North and South of zone 2. Incidence angles are in general higher than those of the zone 2 and T0 is generally lower. Spectral ratios show amplifications comparable or lower than those in zone 2. Border of the Northern part of this zone is quite uncertain.

zone 4

This zone corresponds to “Serraloggia – Spina” and is characterized by spectral ratios which show relevant amplifications, in some cases up to 6 in the interval between 2 and 4 Hz and a little lower but over a wider frequency range in other sites. A significant feature of this zone is the considerable variability of the expected ground motion as evidenced by the difference in the spectral ratios between stations located at less than 200 meters. Incidence angles are the lowest in the Fabriano area. It is possible to delimit such zone only to East at the boundary with zone 2 on the base of T0 map.

As already mentioned, the results of weak motion processing, synthesized in the zonation map, and the damage distribution are generally in a good accordance; the damage, indeed, is concentrated in two suburbs, Borgo and Serraloggia-La Spina (zone 1 and 2) where spectral ratios show the highest amplifications.

Figure 7 Site effects zonation map of Fabriano (Tento et al., 2001)

Comparison between H/V results and geology

Figure 8 shows a map of Fabriano basin depth obtained using borehole data. The basin has an ellipsoidal shape with the major axis elongated parallel to the axis of the valley. Its maximum depth reaches about 30 meters below the eastern part of the historical center.

Figure 9 H/V ratios at some station along a profile perpendicular to the axis of the valley (Tento et al., 2001)

Figure 9 represents noise H/V ratios on some stations located along a NW-SE direction across the Fabriano basin and a T0 profile (upper right panel). The coherence between the deepest portions of the basin and the stations with the lowest values of the fundamental period is fairly good as well as the higher values of fundamental period observable on the two sides of the valley (stations GE2, Garibaldi and Poio; see also Figure 6).

Comparisons between damage distribution and H/V results

The comparison between damage grade (defined according to EMS 98) and the fundamental period of soil is represented in figure 10. At a glance, this figure does not show a clear correspondence between the map of the fundamental period and the damage; on the contrary the pattern of the underlying grid appears to be correlated with the one represented in figure 8, confirming a conformity between the geological properties of the basin and noise H/V processing results.

The four panels of Figure 11 represent for each damage grade, the distribution of the soil period under the analyzed buildings. For instance, the bottom panel (EMS98 damage grade 0) shows an almost uniform distribution of the number of buildings for each interval of the soil fundamental period (all the histograms are normalized to the total number of buildings surveyed). As already observed for figure 10, figure 11 does not show any clear evidence of correlation between soil period and damage.

Figure 10 Comparison between the spatial distribution of deposit fundamental period (orange scale grid) and the spatial distribution of damage - EMS 98 (dots of different colors) for Fabriano.

Figure 11 The four panels show the distribution for each damage grade (according to the EMS98 scale), of the soil period under the analyzed buildings. Each green circle represents a single building. Histograms are normalized in respect to the total number of buildings surveyed.


Using damage database, geological information, weak motion recordings and noise data collected in the framework of the Umbria-Marche microzonation project, we compared the results of noise H/V processing with geological and damage distribution. We found that the fundamental period of the soil as obtained by noise H/V is in agreement with the map of the basin depth obtained using boreholes information (see Figures 8, 9 and 10). However, the areal distribution of damage does not show a clear correlation to the fundamental period of the deposit pattern (Figure 10). Moreover no clear correlation, as it appears from figure 11, was observed between the damage grade and soil fundamental period.


  1. Amato, A., A. Azzara, C. Chiarabba, G.B. Cimini, M. Cocco, M. Di Bona, L. Margheriti, S. Mazza, F. Mele, G. Selvaggi, A. Basili, E. Boschi, F. Courboulex, A. Dechamps, S. Gaffet, G. Bittarelli, L. Chiaraluce, D. Piccinini (1998). The 1997 Umbria-Marche, Italy, earthquake sequence: A first look at the main shocks and aftershocks. Geophysical Research Letters, 25(15), 2861-2864.

  2. Castelli, V., G. Monachesi (2001). Seismic history and historical earthquake scenario for the town of Fabriano (Central Italy). Italian Geotechnical Journal, XXXV(2), 36-46.

  3. Crespellani, T., B. Ciulli, C. Madiai, G. Vannucchi (2001). Dynamic geotechnical testing and seismic response analysis at Fabriano, Italy. Italian Geotechnical Journal, XXXV(2), 146-158.
  4. Dolce, M. and A. Larotonda (2001). Earthquake’s effects on the Fabriano, Nocera Umbra and Sellano’s buildings. Italian Geotechnical Journal, XXXV(2), 10-19.

  5. de Franco R., Boniolo G., Corsi A., Daminelli R., Franceschina G.L., Maistrello M., Morrone A., Pagani M., Tento A., Pierni G., Stoppoloni R., Tiberi P., Michelini A., Govoni A., Duri G., Ponton F., Romanelli F., Cattaneo M., Augliera P., Bindi D., Carenzo G., Eva E., Lanza V., Parolai S., Pasta M., Spallarossa D., Zunino E., Gorini A., Marcucci S., Marsan P., Milana G., Zambonelli E., Mucciarelli M., Monachesi G. (2000) - Registrazioni velocimetriche. In Marcellini A., Tiberi P. (eds.) "La microzonazione sismica di Fabriano", Biemmegraf, Piediripa di Macerata, pp.149-190 (in Italian).

  6. Franceschina G.L. (2000) - CD-ROM delle registrazioni velocimetriche delle repliche effettuate dalla rete installata a Fabriano . In Marcellini A., Tiberi P. (eds.) "La microzonazione sismica di Fabriano", Biemmegraf, Piediripa di Macerata, pp.257-291 (in Italian).

  7. Grünthal, G. (editor) (1998). European Macroseismic Scale 1998. Cahiers du Centre Europeen de Geodynamique et Sismologie, vol 15.

  8. Marcellini, A., P.Tiberi (eds.) (2000). La Microzonazione sismica di Fabriano. Biemmegraf, Piediripa di Macerata.

  9. Marcellini, A., R. Daminelli, A. Tento, G.L. Franceschina,M. Pagani (2001). The Umbria-Marche Microzonation Project: outline of the project and the example of Fabriano results. Italian Geotechnical Journal, XXXV(2), 28-35.

  10. Michelini, A., A. Govoni (2001). Site amplification from earthquake data in Fabriano,Central Italy. Italian Geotechnical Journal, XXXV(2), 118-130
  11. Monachesi, G, M. Stucchi (eds.) (1997). DOM4.1 Un database di osservazioni macrosismiche di terremoti di area italiana, al di sopra della soglia del danno. CNR-GNDT technical report, 2 vols. Available at the website

  12. Parroni, F., G. Scarascia-Mugnozza, P. Traversa, G. Valentini (2001). Engineering geological investigations for the seismic microzonation of Fabriano (Marche, Italy). Italian Geotechnical Journal, XXXV(2), 86-97.

  13. Tento A., R. de Franco , G. Franceschina, M. Pagani (2001). Site effect zonation of the Fabriano municipality. Italian Geotechnical Journal, XXXV(2), 131-145.

1   2   3   4

The database is protected by copyright © 2017
send message

    Main page